For our 5th annual Insight Jam LIVE! Solutions Review editors sourced this resource guide of analytics and data science predictions for 2024 from Insight Jam, its new community of enterprise tech experts.

Note: Analytics and data science predictions are listed in the order we received them.

Analytics and Data Science Predictions from Experts for 2024

Rahul Pradhan, Vice President of Product, Engineering, and Cloud Operations at Couchbase

Real-time data will become the standard for businesses to power generative experiences with AI; Data layers should support both transactional and real-time analytics 

“The explosive growth of generative AI in 2023 will continue strong into 2024. Even more enterprises will integrate generative AI to power real-time data applications and create dynamic and adaptive AI-powered solutions. As AI becomes business critical, organizations need to ensure the data underpinning AI models is grounded in truth and reality by leveraging data that is as fresh as possible.”

“Just like food, gift cards and medicine, data also has an expiration date. For generative AI to truly be effective, accurate and provide contextually relevant results, it needs to be built on real-time, continually updated data. The growing appetite for real-time insights will drive the adoption of technologies that enable real-time data processing and analytics. In 2024 and beyond, businesses will increasingly leverage a data layer that supports both transactional and real-time analytics to make timely decisions and respond to market dynamics instantaneously.”

Expect a paradigm shift from model-centric to data-centric AI

“Data is key in modern-day machine learning, but it needs to be addressed and handled properly in AI projects. Because today’s AI takes a model-centric approach, hundreds of hours are wasted on tuning a model built on low-quality data.”

“As AI models mature, evolve and increase, the focus will shift to bringing models closer to the data rather than the other way around. Data-centric AI will enable organizations to deliver both generative and predictive experiences that are grounded in the freshest data. This will significantly improve the output of the models while reducing hallucinations.”

Multimodal LLMs and databases will enable a new frontier of AI apps across industries

“One of the most exciting trends for 2024 will be the rise of multimodal LLMs. With this emergence, the need for multimodal databases that can store, manage and allow efficient querying across diverse data types has grown. However, the size and complexity of multimodal datasets pose a challenge for traditional databases, which are typically designed to store and query a single type of data, such as text or images. “

“Multimodal databases, on the other hand, are much more versatile and powerful. They represent a natural progression in the evolution of LLMs to incorporate the different aspects of processing and understanding information using multiple modalities such as text, images, audio and video. There will be a number of use cases and industries that will benefit directly from the multimodal approach including healthcare, robotics, e-commerce, education, retail and gaming. Multimodal databases will see significant growth and investments in 2024 and beyond — so businesses can continue to drive AI-powered applications.”

Nima Negahban, CEO and Co-Founder at Kinetica

Generative AI turns its focus towards structured, enterprise data

“Businesses will embrace the use of generative AI for extracting insights from structured numeric data, enhancing generative AI’s conventional applications in producing original content from images, video, text and audio. Generative AI will persist in automating data analysis, streamlining the rapid identification of patterns, anomalies, and trends, particularly in sensor and machine data use cases. This automation will bolster predictive analytics, enabling businesses to proactively respond to changing conditions, optimizing operations, and improving customer experiences.”

English will replace SQL as the lingua-franca of business analysts

“We can anticipate a significant mainstream adoption of language-to-SQL technology, following successful efforts to address its accuracy, performance, and security concerns. Moreover, LLMs for language-to-SQL will move in-database to protect sensitive data when utilizing these LLMs, addressing one of the primary concerns surrounding data privacy and security. The maturation of language-to-SQL technology will open doors to a broader audience, democratizing access to data and database management tools, and furthering the integration of natural language processing into everyday data-related tasks.”

Vasu Sattenapalli, CEO at RightData

NLP-Powered Analytics Will Be the Next Wave of Self Service

“Analytics have been stuck in dashboards, which will no longer be the only way to consume business insights. Voice and Generative AI will enter the analytics space where you can ask questions of your data verbally and get a response back in minutes, if not seconds. Imagine even pulling out your phone with an app specific to your organization’s data and being able to access a world of insights. It’s coming!”

Shawn Rogers, CEO and Fellow at BARC

AI is driving innovation in data management, especially through automation and speed

“Having strength at this core level of your data stack is critical for AI success. NLP and conversational UI’s will open the door for the true democratization of analytics. It’s an exciting time for data and insights.”

Bernie Emsley, CTO at insightsoftware

CTO’s will need to bring even more collaboration and education to the C-suite

“Over the past few years, the CTO role has become the bridge between the tech-savvy and the business-savvy, charged with enabling the right solutions to create the best overall business outcomes. This comes with its communication challenges as the CTO needs to navigate how to translate tech into an ROI for the organization’s board and C-suite. In 2024, the ability to educate their C-level colleagues will become even more important as artificial intelligence (AI) technologies become commonplace. The CTO will not only need to be able to collaborate with the tech side of the business to ensure what is realistically possible in the realm of AI but will need to communicate on a business level its potential – both from employee productivity and product standpoint.”

Strong data engines will make financial data movement possible

“Financial organizations are just starting to realize the potential their data holds, using it for guidance in financial planning and analysis, budgetary planning, and more. However, much of this data is still siloed, and we have reached the point where these organizations have so much of this data, that they need to start thinking about how it can bring value to the company or risk losing their competitive advantage. In 2024, we will see finance organizations seek to classify and harmonize their data across repositories to enable new solutions. In response, data engines, data platforms, and data lakes will be just a few tools that will become crucial to understanding and utilizing such data effectively. As a result, we can expect to see the growth of fintech applications to enable this aggregated data analysis, reporting, and visualization to take place.”

Joy Allardyce, General Manager, Data & Analytics at insightsoftware

A continual shift to cloud resources

“The continued push to re-architect technology landscapes to a cloud/SAAS approach will prevail, and many organizations who’ve made large bets ($1B+) contracts on the cloud will find they can’t innovate fast enough to deliver on those commitments. Some, on the other hand, don’t see it as a migration for cost, but an opportunity to modernize and transform how they use data in their business.”

The rise and adoption of AI

“AI, like all reporting projects, is only as good as the data it has access to and the prompts used to make a request. With the push for AI, many are still stuck getting their data foundations established so that they can take advantage of AI. To avoid pilot purgatory, starting with the outcome (use case) in mind that shows a quick win and demonstrable value vs. a one-off project is key.”

Democratizing data

“While the notion of centralized data management is a trend, the reality is that departments still own their data AND have domain expertise. How organizations can adopt a democratized and open fabric but employ the right data governance strategies to support faster innovation and adoption will be crucial. Doing so will only further support the adoption of AI, which requires strong domain knowledge for value to be truly extracted.”

Andy Oliver, Director of Marketing at CelerData

Java will continue to be used for a great many legacy and even current systems and applications

“Java, though showing its age and looking slower in today’s environments, will continue to be used for a great many legacy and even current systems and applications, regardless of the low level of support and leadership from Oracle*

The challenge with implementing real-time data has been more about storage than anything else. I think in the past people were obsessed with real-time versus batch. Sometimes it seems like a choice between something that’s big enough but too slow vs. something that’s fast enough but too small.

However, real-time and batch will come together, to meet the requirements of user numbers, and we will see more unified analytical database technologies for functions and insights that demand real-time analysis.

Not everything will need to move over to real-time, though – there are plenty of things where there’s no good reason to do it.

I think we’re going to see most of the nonsense shake out from operational AI if it can really learn and stick to core organizational needs, and be deployed responsibly and effectively. That’s where VCs are going to focus in the future, the rest will keep falling by the wayside.”

Casey Ciniello, Product Owner and Marketing Manager at Infragistics

More Businesses Will Rely on Predictive Analytics to Make Decisions in 2024

“Making decisions based on gut instinct is a thing of the past as organizations are fully realizing the power of analytics to make data-driven decisions, evidenced by the number of software platforms incorporating embedded analytics. Analytics will be all encompassing in 2024 as we become reliant on data for everything from everyday business research such as inventory and purchasing to predictive analytics that allow businesses to see into the future. Predictive analytics will drive businesses forward by helping them make informed, data-driven decisions, improve productivity, and increase sales/revenue — rather than merely reacting in response to events that have already taken place.”

Justin Borgman, Co-Founder and CEO at Starburst

Two hot topics, data products & data sharing, will converge in 2024

“Data sharing was already on the rise as companies sought to uncover monetization opportunities, but a refined method to curate the shared experience was still missing. As the lasting legacy of data mesh hype, data products will emerge as that method. Incorporating Gen AI features to streamline data product creation and enable seamless sharing of these products marks the pivotal trifecta moving data value realization forward.”

Mike Carpenter, VC Advisor for Lightspeed Venture Partners

AI to Drive Real-Time Intelligence and Decision Making

“Next year will be foundational for the next phase of AI. We’ll see a number of new innovations for AI, but we’re still years away from the application of bigger AI use cases. The current environment is making it easy for startups to build and prepare for the next hype cycle of AI. That said, 2024 is going to be the year of chasing profitability. Due to this, the most important trend in 2024 will be the use of AI to drive real-time intelligence and decision-making. This will ultimately revolutionize go-to-market strategies, derisk investments, and increase bottom-line value.”

Brian Peterson, Co-Founder and Chief Technology Officer at Dialpad

Influx of data talent/AI skills 

“As businesses continue to embrace AI, we’re going to see not only an increase in productivity but also an increase in the need for data talent. From data scientists to data analysts, this knowledge will be necessary in order to sort through all the data needed to train these AI models. While recent AI advancements are helping people comb through data faster, there will always be a need for human oversight – employees who can review and organize data in a way that’s helpful for each model will be a competitive advantage. Companies will continue looking to hire more data-specific specialists to help them develop and maintain their AI offerings. And those who can’t hire and retain top talent  – or don’t have the relevant data to train to begin with – won’t be able to compete. 

Just like we all had to learn how to incorporate computers into our jobs years ago, non-technical employees will now have to learn how to use and master AI tools in their jobs. And, just like with the computer, I don’t believe AI will eliminate jobs, more so that it will shift job functions around the use of the technology. It will make everyone faster at their jobs, and will pose a disadvantage to those who don’t learn how to use it. ”

The commoditization of data to train AI

“As specialized AI models become more prevalent, the proprietary data used to train and refine them will be critical. For this reason, we’re going to see an explosion of data commoditization across all industries. Companies that collect data that could be used to train chatbots, take Reddit for example, sit on an immensely valuable resource. Companies will start competitively pricing and selling this data.” 

Wayne Eckerson, President at Eckerson Group

“Within five years, most large companies will implement a data product platform (DPP), otherwise known as an internal data marketplace, to facilitate the publication, sharing, consumption, and distribution of data products.”

Helena Schwenk, VP, Chief Data & Analytics Officer at Exasol

FinOps becomes a business priority, as CIOs analyze price / performance across the tech stack

“Last year, we predicted that CFOs would become more cloud-savvy amidst recession fears, and we watched this unfold as organizations shifted to a “do more with less” mentality. In 2024, FinOps practices the financial governance of cloud IT operations, as the business takes aim at preventing unpredictable, sometimes chaotic, cloud spend and gains assurance from the CIO that cloud investments are aligned with business objectives.

As IT budgetary headwinds prevail, the ability to save on cloud spend represents a real opportunity for cost optimization for the CIO. One of the most important metrics for achieving this goal is price/performance, as it provides a comparative gauge of resource efficiency in the data tech stack. Given most FinOps practices are immature, we expect CIOs to spearhead these efforts and start to perform regular price/performance reviews. 

FinOps will become even more important against the backdrop of organizations reporting on ESG and sustainability initiatives. Beyond its role in forecasting, monitoring, and optimizing resource usage, FinOps practices will become more integral to driving carbon efficiencies to align with the sustainability goals of the organization.” 

AI governance becomes C-level imperative, causing CDOs to reach their breaking point

“The practice of AI governance will become a C-level imperative as businesses seek to leverage the game-changing opportunities it presents while balancing responsible and compliant use. This challenge is further emphasized by the emergence of generative AI, adding complexity to the landscape. 

AI governance is a collective effort, demanding collaborative efforts across functions to address the ethical, legal, social, and operational implications of AI. Nonetheless, for CDOs, the responsibility squarely rests on their shoulders. The impending introduction of new AI regulations adds an additional layer of complexity, as CDOs grapple with an evolving regulatory landscape that threatens substantial fines for non-compliance, potentially costing millions.

This pressure will push certain CDOs to their breaking point. For others, it will underscore the importance of establishing a fully-resourced AI governance capability, coupled with C-level oversight. This strategic approach not only addresses immediate challenges, but strengthens the overall case for proactive and well-supported AI governance going forward.”

Florian Wenzel, Global Head of Solution Engineering at Exasol

Expect AI backlash, as organizations waste more time and money trying to ‘get it right’

“As organizations dive deeper into AI, experimentation is bound to be a key theme in the first half of 2024. Those responsible for AI implementation must lead with a mindset of “try fast, fail fast,” but too often, these roles need to understand the variables they are targeting, do not have clear expected outcomes, and struggle to ask the right questions of AI. The most successful organizations will fail fast and quickly rebound from lessons learned. Enterprises should anticipate spending extra time and money on AI experimentation, given that most of these practices are not rooted in a scientific approach. At the end of the year, clear winners of AI will emerge if the right conclusions are drawn.

With failure also comes greater questioning around the data fueling AI’s potential. For example, data analysts and C-suite leaders will both raise questions such as: How clean is the data we’re using? What’s our legal right to this data, specifically if used in any new models? What about our customers’ legal rights? With any new technology comes greater questioning, and in turn, more involvement across the entire enterprise.”

Nick Elprin, Co-Founder and CEO at Domino Data Lab

An army of smaller, specialized Large Language Models will triumph over giant general ones

“As we saw during the era of “big data” — bigger is rarely better. Models will “win” based not on how many parameters they have, but based on their effectiveness on domain-specific tasks and their efficiency. Rather than having one or two mega-models to rule them all, companies will have their own portfolio of focused models, each fine-tuned for a specific task and minimally sized to reduce compute costs and boost performance.”

Generative AI will unlock the value and risks hidden in unstructured enterprise data

“Unstructured data — primarily internal document repositories — will become an urgent focus for enterprise IT and data governance teams. These repositories of content have barely been used in operational systems and traditional predictive models to date, so they’ve been off the radar of data and governance teams. GenAI-based chat bots and fine-tuned foundation models will unlock a host of new applications of this data, but will also make governance critical. Companies who have rushed to develop GenAI use cases without having implemented the necessary processes and platforms for governing the data and GenAI models will find their projects trapped in PoC purgatory, or worse. These new requirements will give rise to specialized tools and technology for governing unstructured data sources.”

Kjell Carlsson, Head of Data Science Strategy and Evangelism at Domino Data Lab

Predictive AI Strikes Back: Generative AI sparks a traditional AI revolution

“The new hope around GenAI drives interest, investment, and initiatives in all forms of AI. However, the paucity of established GenAI use cases, and lack of maturity in operationalizing GenAI means that successful teams will allocate more than 90% of their time to traditional ML use cases that, despite the clear ROI, had hitherto lacked the organizational will.”

GPUs and GenAI Infrastructure Go Bust

“Gone are the days when you had to beg, borrow and steal GPUs for GenAI. The combination of a shift from giant, generic LLMs to smaller, specialized models, plus increased competition in infrastructure and also quickly ramping production of new chips accelerated for training and inferencing deep learning models together mean that scarcity is a thing of the past. However, investors don’t need to worry in 2024, as the market won’t collapse for at least another year.”

Forget Prompt Engineer, LLM Engineer is the Least Sexy, but Best Paid, Profession

“Everyone will need to know the basics of prompt engineering, but it is only valuable in combination with domain expertise. Thus the profession of “Prompt Engineer” is a dud, destined, where it persists, to be outsourced to low-wage locations. In contrast, as GenAI use cases move from PoC to production, the ability to operationalize GenAI models and their pipelines becomes the most valuable skill in the industry. It may be an exercise in frustration since most will have to use the immature and unreliable ecosystem of GenAI point solutions, but the data scientists and ML engineers who make the switch will be well rewarded.”

GenAI Kills Quantum and Blockchain

“The unstoppable combination of GenAI and Quantum Computing, or GenAI and Blockchain? Not! GenAI will    be stealing all the talent and investment from Quantum and blockchain, kicking quantum even further into the distant future and leaving blockchain stuck in its existing use cases of fraud and criminal financing. Sure, there will be plenty of projects that continue to explore the intersection of the different technologies, but how many of them are just a way for researchers to switch careers into GenAI and blockchain/quantum startups to claw back some of their funding?”

Arina Curtis, CEO and Co-Founder at DataGPT

Data and Business Teams Will Lock Horns Onboarding AI Products

While business user demand for AI products like ChatGPT has already taken off, data teams will still impose a huge checklist before allowing access to corporate data. This tail wagging the dog scenario may be a forcing function to strike a balance, and adoption could come sooner than later as AI proves itself as reliable and secure.”

Businesses Big and Small Will Prioritize Clean Data Sets

“As companies realize the power of AI-driven data analysis, they’ll want to jump on the bandwagon – but won’t get far without consolidated, clean data sets, as the effectiveness of AI algorithms is heavily dependent on the quality and cleanliness of data. Clean data sets will serve as the foundation for successful AI implementation, enabling businesses to derive valuable insights and stay competitive.”

Doug Kimball, CMO at Ontotext

Shift from How to Why: Enter the Year of Outcome-based Decision Making

“In 2024, data management conversations will experience a transformative shift and pivot from “how” to “why.” Rather than focusing on technical requirements, discussions next year will shift to a greater emphasis on the “why” and the strategic value data can bring to the business. Manufacturers recognize that data, once viewed as a technical asset, is a major driver of business success. Solution providers that deal with these needs are also seeing this change, and would be wise to respond accordingly.

In the coming year, data strategy and planning will increasingly revolve around outcomes and the value/benefit of effective data management, as leaders better understand the key role data plays in achieving overarching business objectives. Manufacturers will also reflect on their technology spend particularly those that have yielded questionable results or none at all. Instead of technical deep dives into intricacies like data storage and processing, crafting comprehensive data strategies that drive lasting results will be the priority.

Next year, manufacturers will move beyond technical deep-dives and focus on the big picture. This strategic shift signals a major change in the data management mindset for 2024 and beyond, ideally aligning technology with the broader objectives of the business such as driving growth, enhancing customer experiences, and guiding informed decision-making.”

Christian Buckner, SVP, Data Analytics and IoT at Altair

AI Fuels the Rise of DIY Physics-based Simulation 

“The rapidly growing interaction between Data/AI and simulation will speed up the use of physics-based simulations and extend its capabilities to more non-expert users.”

Mark Do Couto, SVP, Data Analytics at Altair

AI Will Need to Explain Itself

“Users will demand a more transparent understanding of their AI journey with “Explainable AI” and a way to show that all steps meet governance and compliance regulations. The White House’s recent executive order on artificial intelligence will put heightened pressure on organizations to demonstrate they are adhering to new standards on cybersecurity, consumer data privacy, bias and discrimination.”

Molham Aref,  Founder and CEO at RelationalAI

2024: the Rise of the Data Cloud to Advance AI and Analytics 

“While data clouds are not new, I believe there will be a continued emergence and a clear distinction made between data clouds and compute clouds in 2024. With compute clouds like AWS or Azure, we have had to assemble and stitch together all the components needed to work with AI. So with data clouds, like Snowflake or Microsoft Fabric, users have it all pre-packaged together in a single platform, making it much easier to run analytics on data needed to build AI systems. The rise of the data clouds will offer a better starting point for data analytics and Artificial Intelligence (AI) and Machine Learning (ML).”

Dhruba Borthakur, Co-Founder and CTO at Rockset

In 2024, Enterprises Get A Double Whammy from Real-Time and AI – More Cost Savings and Competitive Intelligence 

“AI-powered real-time data analytics will give enterprises far greater cost savings and competitive intelligence than before by way of automation, and enable software engineers to move faster within the organization. Insurance companies, for example, have terabytes and terabytes of data stored in their databases, things like documentation if you buy a new house and documentation if you rent. 

With AI, in 2024, we will be able to process these documents in real-time and also get good intelligence from this dataset without having to code custom models. Until now, a software engineer was needed to write code to parse these documents, then write more code to extract out the keywords or the values, and then put it into a database and query to generate actionable insights. The cost savings to enterprises will be huge because thanks to real-time AI, companies won’t have to employ a lot of staff to get competitive value out of data.”

The Rise of the Machines Powered by Real-Time Data and AI Intelligence

“In 2024, the rise of the machines will be far greater than in the past as data is becoming more and more “real-time” and the trajectory of AI continues to skyrocket. The combination of real-time data and AI make machines come to life as machines start to process data in real-time and make automatic decisions!”

Register for Insight Jam (free) to gain exclusive access to best practices resources, DEMO SLAM, leading enterprise tech experts, and more!

The post 36 Analytics & Data Science Predictions from 22 Experts for 2024 appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by our enterprise tech thought leader community. In this feature, KNIME‘s Rosaria Silipo offers commentary on the changing composition of data science teams.

The era of the lone data scientist is over. 

As business leaders look to capitalize on the potential of data science, many have turned their attention toward finding more specialized talent — and empowering more generalists to complete what used to be complex data science tasks. Increasingly, leaders are recognizing the importance of developing well-rounded teams that can ideate, put new machine learning models into production, and integrate pre-existing machine learning models.  

From a business perspective, organizations need the ability to recruit specialized talent to capitalize on emerging data science trends — especially those related to the rise of new AI tools. And potential job-seekers need to know which skills these businesses are looking for, as well as how to market themselves as specialists and showcase their unique areas of expertise.  

The Changing Composition of Data Science Teams

The New Data Science Landscape 

Just a few short years ago, most businesses assigned a single person to handle all data science-related tasks: the data scientist. This data scientist was tasked with overseeing data acquisition, data storage, data cleaning, model training, model tuning, productionization, and more. Many businesses had to experience less-than-ideal outcomes, such as delayed project timelines and subpar model performance, before they realized that they could no longer rely on a single data scientist to handle all of these tasks. 

 The rise of new AI applications has also demonstrated the ineffectiveness of this generalist approach. Given the complexity of developing AI applications from scratch, many businesses have shifted their focus toward adopting and tinkering with pre-existing models. A greater focus on integration means that, in many cases, engineers are a better fit for these roles than true data scientists. 

As a result, businesses have shifted their hiring focus to more specialized data science roles, including: 

• Data engineers: Data engineers feed the pipeline with massive amounts of historical data and prepare it for analysis. They do this by subjecting data to various quality control measures and storing it in a data warehouse or lake. Without data engineers, data scientists would have to use low quality data to train their models.  
• Data analysts: Data analysts are typically tasked with reports and visualization of trends and KPIs. These reports serve to check the status quo of the data, or of the business as a whole. 
• Machine learning engineer. These engineers are the most recent addition to businesses’ data science teams. Situated somewhere between a data engineer and a data scientist, ML engineers oversee the tuning and productionization of machine learning models.  
• Data scientists. Data scientists do still play a role in modern data science teams. However, their role is now more confined to the creation and training of machine learning models, and they must rely on collaboration with more engineering-like professional figures. 

A renewed focus on roles outside of the traditional data scientist gives businesses the versatility to harness their data more effectively.  

Specialization is Key for Data Science Teams  

From the employee perspective, job-seekers must adjust to these new hiring preferences to land data science jobs. With integration and engineering skills more valuable than ever, job-seekers should develop or deepen their expertise in these areas, taking a more engineer-like approach instead of a solely creative one.  

In today’s competitive hiring environment, practical experience is also a must. It’s not enough to simply possess technical skills. Tech companies are looking for proven experience in application building and model training — as well as softer skills like project organization and result presentation. And while a breadth of experience is beneficial, candidates who specifically market themselves as a data analyst or data engineer may find it easier to attract the attention of employers. 

Modern data science teams require extensive engineering, analytics, and integration knowledge to complete projects. It’s clear that specialization is the only way to achieve this balance. This trend toward forming well-rounded teams capable of innovation, deployment, and integration will only continue as data science leaders work to develop a deeper understanding of their industry. 

The post The Changing Composition of Data Science Teams appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise tech. In this feature, Kyvos Insights’ Pratik Jain offers key advantages of a semantic layer inside the modern data stack.

A significant shift is happening in how data is seen by businesses today. There is an increasing recognition of its formidable potential for business transformation. Data is increasingly perceived as a potent resource capable of driving innovations, efficiency, product enhancements, customer engagement, and overall strategic decision-making.

This shift is leading a notable change in who drives data requirements. Earlier, the impetus for data needs predominantly came from analysts or IT teams, but now business users are taking a lead role. They want to ensure that the insights that they need from data are readily available, and tailored to their specific needs.

Business users, however, lack the technical expertise to work with complex data schemas, SQL queries and intricate data structures. What they need is a user-friendly interface and an intuitive language to interact with data, and this is where the concept of a semantic layer comes into play. It bridges the gap between the complexity of underlying data and the accessibility that business users require, enabling them to extract meaningful insights without being data experts themselves.

Quoting from Wikipedia, a semantic layer is a “business representation of corporate data that helps end users access data autonomously using common business terms”. The semantic layer’s core purpose is to enhance data’s utility for the business. Acting as an abstraction layer, it maps the source data into familiar business views using dimensions, measures and hierarchies that business users are familiar with, like products, sales, territories, time period, customer ID etc.

Also called the metric layer, a semantic layer is sandwiched between the presentation layer (BI, analytics tools, data science tools) and the data warehouse or data lake. It has been called the “foundational plank” in a modern data stack that makes it functional, practical, efficient and scalable. The advantages of a semantic layer are manifold:

Advantages of a Semantic Layer

Self-Service

A semantic layer allows business users to independently explore and derive granular insights from the data with minimal or no reliance on IT and data experts. They can create custom reports, perform ad-hoc analyses and gain valuable insights on their own. Organizations can significantly reduce bottlenecks in the data/ insight delivery process by eliminating the need for IT or data experts to generate reports or answer data-related queries.

With self-service analytics through a semantic layer, business users can swiftly access and explore the data leading to faster insights. They can ask questions, refine queries and investigate data anomalies in real-time, enabling quicker responses to changing market conditions or emerging opportunities

Single Source of Truth

Various departments in an organization tend use different BI tools – each having their own metric definitions. For example, a week may be defined as starting Monday or Sunday; time as local or GMT; leading to different versions of reports from same source data.

A semantic layer provides a single unified set of metrics within a data stack across the organization that are shared across various BI tools. By harmonizing and standardizing the metric definitions across the company, it fosters a single source of truth.

Departments can continue to use their preferred BI tool to access and interpret data. This promotes a common and consistent understanding of key business performance indicators across the organization and raises the trust in data-driven decisions.

Improved Performance

An organization’s data and analytics requirements continues to evolve and expand over time. Modern data stacks should be designed to be highly scalable so that they stay current. A semantic layer in the stack empowers exceptional performance for enterprise-wide analytics, with minimal degradation as the scale increases.

It pre-aggregates data to optimize performance across numerous dimensions and measures, without the need for data movement. Even as your data workloads grow exponentially, users can obtain instant insights and answers to their queries. Besides a unified view, scalability and speed are critical in a data-driven business as agility and responsiveness are key to maintaining a competitive edge.

Unified and Efficient Data Modeling

A Semantic layer creates a unified data model across all data sources that may reside in disparate systems across departments, on-prem or cloud, as well as geo locations. Irrespective of BI tools or storage platforms, it presents a consistent view of enterprise data.

Data modeling tasks are simplified with a Semantic layer with these added benefits:

• It functions on top of diverse data sources providing virtualization and federation capabilities, enabling business users to access data from multiple sources as if it were centralized.
• Users can define intricate calculations and express complex business logic facilitating the extraction of deeper insights from the data.
• The layer supports the addition of new data sources without disrupting the existing business data view, ensuring scalability and adaptability.
• It optimizes data access, eliminating redundancy and latency, resulting in seamless and efficient data retrieval.

Data Security and Governance: By acting as a gatekeeper between the enterprise data storage system and BI tools, a semantic layer incorporates a formidable mechanism for enforcing security measures. It allows for precise and granular control over who can access what data, implementing row and column-level security measures, at both group and user levels.

Sensitive data remains protected and is accessible only to authorized individuals. This meticulous control contributes to the organization’s overall data governance strategy, enhancing data quality and integrity while preserving data privacy and compliance.

In summary, the implementation of a semantic layer not only simplifies data analytics but also enhances data-driven decision-making by providing a unified, efficient and scalable approach and accessibility of data insights across the enterprise.

The post Advantages of a Semantic Layer in a Modern Data Stack appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise technology. In this feature, dotData CEO Ryohei Fujimaki offers commentary on how new predictive analytics techniques are leading in fraud prevention.

For years we’ve known that fraud is so widespread that it affects national GDPs and that it increases with financial downturns. As all industries and sectors embrace digital transformation, the digital attack surface expands, presenting new fraud opportunities for cybercriminals. A study from Juniper Research found that the online payment fraud losses between 2023 and 2027 will exceed $343 billion.

New machine learning (ML) models, AI applications, and predictive analytics techniques are being leveraged to combat the rising criminal trend, taking offensive approaches to make an impact. From identifying patterns and anomalies to detecting risks in large and complex data sets, predictive analytics can shut down fraud before it happens.

Predictive Analytics Fraud Prevention

Feature Engineering for Fraud Detection

One of the critical challenges in fraud detection is to extract relevant and informative features from the data that can capture the characteristics and behaviors of fraudsters. 

Feature engineering is often manual and time-consuming, requiring domain knowledge and expertise. However, some recent advances in ML have enabled automated feature engineering methods that can reduce human effort and improve the quality of the features. Let’s look at some examples. 

AutoML 

AutoML is a framework that automates the end-to-end process of ML model development, including data preprocessing, feature engineering, model selection, hyperparameter tuning, and model evaluation. AutoML can help find the optimal combination of features and models for a given problem without human intervention. 

Deep Learning 

Deep learning is also being used in fraud detection. As a branch of ML that uses artificial neural networks (ANNs), the technique can be used to learn complex and nonlinear patterns from the data. Deep learning can perform feature engineering implicitly by learning high-level representations or embeddings from the raw data, such as images, text, or audio. Deep learning can also perform feature engineering explicitly by using techniques such as autoencoders or generative adversarial networks (GANs) to create new features from the data.

Reinforcement Learning 

Reinforcement learning can perform feature engineering using techniques such as policy gradients or deep Q-networks (DQNs) to learn optimal policies or strategies for feature selection or generation.

Model Building for Fraud Detection

Building accurate and robust ML models capable of handling the complexity and dynamics of historical and live fraud detection scenarios is exceptionally challenging. These models must deal with big data, imbalanced data, drift, and adversarial attacks. 

As fraud patterns can change over time due to changes in customer behavior, business environment, or fraudster tactics, models must be updated and well-maintained. They also need to be resilient to criminal interference and manipulation of data,. adding layers of security where possible to recognize evasion techniques. 

To address these issues, some of the latest techniques for building predictive models for fraud detection include ensemble learning, active learning, semi-supervised learning, and others. 

Ensemble learning is a technique that combines multiple models to create a more robust model. This strategy can improve the performance and stability of fraud detection models by reducing variance, bias, or overfitting. Ensemble learning can combine different models or algorithms, such as bagging, boosting, stacking, or voting.

On the other hand, active learning is used to select the most informative samples from a large pool of unlabeled data for human annotation, while semi-supervised learning leverages labeled and unlabeled data to train a model. 

If a company struggles with data scarcity, semi-supervised learning can be used as it utilizes the abundant unlabeled data to work around the problem. Semi-supervised learning can use self-training, co-training, or graph-based methods to propagate labeled and unlabeled labels.

Deep learning is a technique that uses multiple layers of neural networks to learn complex and nonlinear features from the data. 

Deep learning can capture fraud data’s high-dimensional and heterogeneous nature by extracting abstract and meaningful representations. This approach can use architectures such as autoencoders, convolutional neural networks, recurrent neural networks, or attention mechanisms to model different types of fraud data.

Different Sectors, Different Approaches to Fraud

It is also important to note that fraud techniques vary depending on their target sector. Each industry must adapt its countermeasures accordingly. 

Banks and digital finance are among the most likely to be exposed to fraud attacks, as they involve high-value transactions, sensitive data, and multiple channels and parties. Cybercriminals will adopt a wide array of techniques to launch fraud campaigns in this sector. These include deceiving customers or employees through mass phishing, whaling, spear phishing, identity theft, account takeover, card skimming, money laundering, and others. Predictive analytics can help banks detect and prevent fraud using different concepts. 

One of the most effective methods to detect banking fraud is generating scoring transactions based on risk level. ML models, trained on historical data and equipped with real-time features—amount, location, device, behavior, etc.—can generate very high-level risk scores, which, in turn, can be used to flag and shut down fraud attacks. Scoring has become a mainstream fraud security feature. Companies like IBM, through Watson Studio, provide a platform that supports visual programming and deep learning to develop and deploy ML models for fraud scoring.

Another way to go is to segment customers using profile data and behavior. In this strategy, ML models using techniques like clustering or classification identify normal and abnormal patterns and flag suspicious activities. Mastercard has been using this technique since 2019, segmenting customers into groups based on their spending habits and preferences to monitor their transactions for deviations from their usual patterns.

Natural language processing (NLP) and natural language generation (NLG) techniques can also be leveraged in ML models to generate alerts and recommendations and inform stakeholders of possible or existing fraud. The most considerable disruption in this area is ChatGPT, a natural language processing (NLP) technology that can be used as a novel tool in fraud detection and investigation. NLP can identify patterns and anomalies, enabling fraud investigators to streamline anti-fraud technologies. 

But other sectors are not immune to fraud. In e-commerce, where online transactions are often anonymous, cross-border, and subject to chargebacks, criminals exploit online platforms and merchants using fake accounts, stolen credit cards, and refund abuse.

ML models that leverage biometric data (such as face or voice recognition), behavioral data (such as keystrokes or mouse movements), or device data (such as IP address or browser fingerprint) can be used to verify the identity and authenticity of customers. 

In the insurance sector, ML developers need to adapt to combat complex, subjective, or delayed claims. Claims can be inflated or falsified, accidents can be staged, injuries exaggerated, and documents fabricated.

The post New Predictive Analytics Techniques Leading in Fraud Prevention appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise technology. In this feature, Prophix Chief Customer Innovation Officer Susan Gershman offers commentary on the self-service BI advantage when it comes to insight-driven finance.

The focus for businesses has shifted from whether they are using data analytics to how deeply these analytics are integrated into companies’ operations. Data analytics have extended beyond the confines of technical teams and IT departments. Increasingly, various areas throughout organizations are benefiting from data-driven decision making and analytical insights. And as a result, companies are in search of solutions that can deliver these capabilities more broadly and instinctively, reaching into new departments and involving individuals without specialized technical skills. This is where self-service business intelligence (BI) plays a pivotal role.

Self-service BI is an approach to data analytics that enables users to analyze and explore data sets without the aid of an organization’s data scientists or BI team. These tools make it easy for users without a background in traditional analytics to gain insights from their organization’s data. While self-service BI can aid departments across organizations, finance teams are especially benefitting from the speed, efficiency, and agility of self-service BI. In today’s ever-evolving business landscape, finance professionals are focusing on ways to keep up with the pace of change and level-up their strategic value. The ability to rapidly gain financial intelligence and apply that insight to drive significant business results is key to their success.

Self-Service BI in Finance

The New Era of Finance

Finance teams tend to be an independent, self-sufficient group. They want to be able to do their own analysis and use their own tools, without relying on outside help. These teams are also typically equipped with a baseline analytical skill set — most finance professionals are comfortable using Excel or similar tools. But now that the pressure is on for finance teams to move beyond spreadsheets, they’re looking for more advanced software that can provide higher-level insights…so long as they can still navigate the tools without leaning on technical teams. Adopting self-service BI tools within the finance department is a natural transition as teams look to bolster their analytical capabilities to drive business impact.

The finance department needs to move faster than it was ever previously expected. To be most impactful in their roles and make informed decisions quickly, rapid access to data is simply vital. Yet many teams remain bogged down manually gathering details from other departments or waiting days for a BI team to return the report they requested — only for circumstances to change once again, requiring a new report. Instead, they need tools that empower collaboration, efficiency, and improved decision-making.

The Benefits of Self-Service BI in the Finance Department

Self-service BI makes accessing critical data insights faster, meaning decision making can get done faster, too. With self-service BI, everything can be contained in one place, avoiding any unnecessary steps or bottlenecks. Having this single version of the truth is essential to the

finance department for accurate and efficient planning, reporting, consolidation, and account reconciliation. This means that all decision-makers can undertake key financial processes and analysis from one location while maintaining data integrity and accuracy, and reducing time spent on manual input that is inherent with spreadsheets.

As AI continues to improve at seemingly breakneck speed, self-service BI tools are also getting more powerful and easier to use. With natural language processing, users can interact with BI tools as if they are having a natural conversation, asking questions of the data and getting immediate insights. These developments enable finance professionals to interact with the data in an intuitive way, instead of the manual slicing and dicing required when working in a spreadsheet, or even the more traditional analysis methods typically found in classic BI tools.

The most transformative shift in finance over the past few years is the imperative for finance teams to go beyond reporting and metrics to driving business strategy. Automation has done away with a lot of the rote tasks that finance teams have long been known for, leaving room for finance professionals to step up into more strategic roles. The CFO’s role has transformed from budget guardian to leader and business advisor within the organization. Self-service BI gives finance executives the power to identify, monitor, and improve the metrics that are most important to their company’s performance, solidifying their rightful and growing influence in organizational decision-making.

In Closing

Enabling self-service BI requires more than just deploying user-friendly tools. Self-service tools don’t work unless an organization has robust data collection capabilities and the data is kept clean and ready-to-consume. No matter how self-sufficient and technically skilled a finance team is, the organization needs to be fully committed to ensuring these tools are able to be used effectively.

Adoption of self-service BI in finance departments is growing as teams look for efficient and effective solutions that will empower them to be the best they can be in their roles. With these tools, they can rapidly gain the intelligence they need and apply that knowledge to drive positive business results.

The post Insight Driven Finance: The Self-Service BI Advantage appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

The editors at Solutions Review have compiled this list of the best MLOps courses and online training to consider.

MLOps (short for machine learning operations) is the process of taking a model developed in an experimental environment and putting it into a production web system. When an application is ready to be launched, MLOps is coordinated between data science professionals, DevOps and machine learning engineers to transition the algorithm into production. A key characteristic of MLOps is to increase automation and improve the quality of production machine learning while keeping compliance requirements in check.

With this in mind, we’ve compiled this list of the best MLOps courses and online training to consider if you’re looking to grow your data science and machine learning skills for work or play. This is not an exhaustive list, but one that features the best MLOps courses and online training from trusted online platforms. We made sure to mention and link to related courses on each platform that may be worth exploring as well.

The Best MLOps Courses

TITLE: Machine Learning Monitoring Concepts

OUR TAKE: This course, instructed by NannyML co-founder and CEO Hakim Elakhrass, starts with the blueprint of where to begin monitoring in production and how to structure the processes around it. It also covers basic workflow by showing you how to detect the issues, identify root causes, and resolve them with real-world examples.

Platform: DataCamp

Description: Deploying a model in production is just the beginning of the model lifecycle. Even if it performs well during development, it can fail due to continuously changing production data. In this course, you will explore the difficulties of monitoring a model’s performance, especially when there’s no ground truth. Learn all about the challenges of monitoring machine learning models in production, including data and concept drift, and methods to address model degradation.

TITLE: MLOps (Machine Learning Operations) Fundamentals

OUR TAKE: Offered by Google Cloud, this 100 percent online training features flexible deadlines on intermediate-level subject matter. It takes roughly 16 hours to complete.

Platform: Coursera

Description: This course introduces participants to MLOps tools and best practices for deploying, evaluating, monitoring and operating production ML systems on Google Cloud. MLOps is a discipline focused on the deployment, testing, monitoring, and automation of ML systems in production. Machine Learning Engineering professionals use tools for continuous improvement and evaluation of deployed models. They work with (or can be) Data Scientists, who develop models, to enable velocity and rigor in deploying the best performing models.

TITLE: Applied Machine Learning: Foundations

OUR TAKE: Data scientist Derek Jedamski specializes in machine learning and shows students the Python programming language, machine learning techniques and data cleaning examples.

Platform: LinkedIn Learning

Description: In this course, the first installment in the two-part Applied Machine Learning series, instructor Derek Jedamski digs into the foundations of machine learning, from exploratory data analysis to evaluating a model to ensure it generalizes to unseen examples. Instead of zeroing in on any specific machine learning algorithm, Derek focuses on giving you the tools to efficiently solve nearly any kind of machine learning problem.

TITLE: Demistifying Machine Learning Operations (MLOps)

OUR TAKE: This intermediate-level Pluralsight training is just two hours long and will teach you the main concerns and issues to consider while developing machine learning models after deployment.

Platform: Pluralsight

Description: In this course, Demystifying Machine Learning Operations (MLOps), you’ll learn to implement machine learning operations into your machine learning project. First, you’ll explore how to apply machine learning operations (MLOps) practices for your infrastructure. Next, you’ll discover how machine learning operations (MLOps) during model development. Finally, you’ll learn how to apply machine learning operations (MLOps) after model deployment.

TITLE: Become a Machine Learning Engineer for Microsoft Azure Nanodegree

OUR TAKE: Udacity’s nanodegree program takes roughly 3 months to complete (at 5-10 hours per week). Students should bring prior experience with Python, machine learning, and statistics for the greatest chance of success with this module.

Platform: Udacity

Description: In this program, students will enhance their skills by building and deploying sophisticated machine learning solutions using popular open source tools and frameworks, and gain practical experience running complex machine learning tasks using the built-in Azure labs accessible inside the Udacity classroom.

NOW READ: The Best Coursera Machine Learning Training to Consider

Solutions Review participates in affiliate programs. We may make a small commission from products purchased through this resource.

The post The 5 Best MLOps Courses and Online Training for 2023 appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise technology. In this feature, Anaconda‘s co-founder and CEO Peter Wang offers commentary on the top ways generative AI will transform the classroom.

Class is officially in session, and the AI craze is taking schools by storm. Everywhere you look, students and teachers are turning to tools like ChatGPT, and the attitude among many is that this new technology has fundamentally changed education forever – and it’s here to stay.

While the sci-fi vision of a robot in the classroom may be difficult to envision now, AI is already in the classroom, and it’s helping to level the playing field and enhance educational outcomes for students of all backgrounds.

The possibilities that generative AI brings to education will transform the classroom, yet there are also ethical considerations and challenges that must be addressed, like ensuring the quality and accuracy of AI-generated content, navigating privacy concerns, and maintaining a balance between human and AI involvement throughout the learning process. The successful integration of generative AI into the classroom will require careful planning, collaboration, and the ongoing assessment of its impact on learning outcomes and students.

Generative AI’s and the Classroom

Here are ten ways generative AI could transform education and unlock new paths of learning:

1. Customized Learning: Each student understands material differently; there are visual and auditory learners, those that learn by observing, and others by experimenting. AI can create customized, independent learning materials by tailoring lesson content and classroom activities to individual students’ learning styles, preferences, and pace, successfully differentiating content for all types of learners.
2. Content Creation: Generative AI can assist teachers in generating educational content such as tests, quizzes, and assignments. In turn, teachers can save time and focus on more interactive and personalized teaching methods.
3. Automated Grading: Teachers spend five hours of grading each week on average – time that could be spent making deeper connections with students or ideating on new lesson plans. Introducing generative AI to their workflows can save time on the repetitive, manual aspects of grading by automating the process for certain assignments like homework or tests.
4. Stronger Student / Teacher Connections: With time-savings a key benefit to incorporating generative AI to the school environment, educators can more easily take a step back from preparing materials and grading to focus on putting the human touch on curriculum planning, the support of social learning, and getting to know students on a more personal level. With this, we can expect the role of the teacher to transition to a coach or mentor in the coming years.
5. Tutoring and Extra Support: Generative AI can function as a tutoring system, providing students with instant feedback, explanations, and additional resources as they work through new concepts. As these AI platforms and large language models (LLMs) develop, we can expect to see customized platforms designed specifically for education that is even better for students and teachers.
6. Language Learning and Translation: Language models powered by AI can assist in overcoming learning barriers by providing real-time translations, language practice, and context-based language acquisition.
7. Collaborative Learning: Facilitating collaborative learning, today’s emerging AI platforms can help students brainstorm ideas, co-create content, and solve problems together, regardless of physical location.
8. Critical Thinking and Problem Solving: AI-generated study materials can engage students in critical thinking and problem-solving exercises, rather than just giving them the answer to their question or problem at hand.
9. Assistive Technology: For students with disabilities, AI can generate accessible content like providing real-time closed captions or adapting materials to accommodate diverse learning needs or pacing.
10. Simulation and Virtual Reality: As mentioned above, there are all types of learners – for some, traditional teaching methods can be challenging, and visual instruction is needed for better understanding. Generative AI can create realistic simulations and virtual reality environments for more complex subjects, enhancing experiential learning.

With these use cases quickly becoming a reality in education, the first and most important step is for school districts and administrators to confront the intricate data challenges already surfacing. This includes preventing biased data from influencing generative AI applications and safeguarding against data breaches and unauthorized data transfers. The emerging data challenges confronting the broader industry will only be amplified within educational systems, necessitating trustworthy data partnerships to instill effective governance over these tools.

Schools are already under immense pressure to keep their data and systems secure. Ransomware has had a devastating impact on school districts across the US with 80% of school IT leaders saying they had been the victim of a ransomware attack. LLMs and AI, which require vast amounts of data, could make schools an even bigger target and underscores the importance of strong cybersecurity practices.

Educators and school administrators need to take into account several crucial considerations as they begin their AI journey. In the short term, there will be a growing need to train more educators in effectively utilizing generative AI tools to harness its potential for automating tasks like creating educational materials, all while addressing issues such as academic dishonesty. Additionally, students will require training on leveraging AI as a self-learning resource, fostering their innate curiosity and enabling them to discern between AI-generated content and reliable sources of information.

While we must tread carefully, AI in the classroom allows educators to focus on what they excel most at: forming connections with their students, teaching core subjects, and inspiring young minds. With this shift, students will have the opportunity to further explore their interests and unleash their creativity. Now, it falls upon school systems and administrators to adapt to this transformation and provide a secure, transformative educational experience.

The post 10 Ways Generative AI Will Transform the Classroom appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise technology. In this feature, Qlik‘s Chief Strategy Officer James Fisher offers commentary on how traditional AI continues to deliver real impact, even in the shadow of buzzy generative AI.

Despite the captivating attention generative AI garners due to its transformative potential, I caution business leaders not to lose sight of traditional AI and ML. I see many excellent use cases involving traditional AI in action each day, extending beyond data scientists and engaging employees who might not be as technical. Many traditional AI methods still hold immense merit and relevance – likely more so even than generative AI will hold in the near term. In fact, McKinsey sees traditional AI continuing to account for the majority of the overall potential value of AI. It makes sense, since traditional AI provides a variety of proven applications including automated insights, predictive modeling, intelligent alerting and natural language processing for tasks like text classification. While an incredible new functionality, generative AI is still an emerging technology that at this time is best suited for content generation and summarization or extending the capabilities of traditional chatbots.

This is why in my final piece in this series, I want to highlight a few of my favorite examples of where traditional AI is delivering measurable value for organizations of various sizes, and across many industries, underscoring the impact it has already amassed in businesses and boardrooms alike.

Under Generative AI’s Shadow, Traditional AI Delivers Impact Now

Traditional AI Helps Meet Sustainability Goals

Perhaps one of the most exciting ways I see traditional AI in action for organizations around the world is what it can do to advance sustainability efforts. C40 is a global network of nearly 100 mayors of the world’s leading cities that are united in action to confront the climate crisis. By applying machine learning to climate datasets – which are often massive and unruly – and combining it with AI tools, C40 can analyze climate trends and emissions data to find patterns that can assist them, and the cities they work for, to take action. Today, 840,000 users across 17,000 municipalities are drawing upon these insights through C40’s knowledge hub. Cities around the world like Cape Town, South Africa and Toronto, Canada have been able to build tangible plans to cut emissions while improving the health of its people.

Traditional AI Uncovers new B2B Revenue Opportunities

In terms of B2B, AI can help across a variety of industries from financial services to retail to technology. In one example, software and services firm for higher education, Gray Associates, Inc., enables its clients to create powerful, data-informed institutional strategies that maximize outcomes for students, schools and other constituencies. To do so, Gray employs AI-powered analytics and automated machine learning to strengthen its location assessment and predictive modeling service. This service guides institutions on large investments – such as opening a new campus – by harnessing internal data and building predictive insights through pulling in big data sources, including US Census data, as well as competitive signals like regional job postings and similar programs from other relevant schools. Gray continues to support many higher education clients with outstanding outcomes – StrataTech Education Group reported revenue of $1.5 million from a new school it opened in Houston following Gray’s advice.

Traditional AI Improves Patient Outcomes & Achieves Significant Healthcare Savings

When it comes to healthcare, there are countless applications in which AI can help reimagine patient and provider experiences. A shining example of an organization driving significant value with traditional AI is non-profit healthcare system Appalachian Regional Healthcare (ARH). Using automated machine learning, ARH is able to identify which of its patients are most susceptible to missing or cancelling their appointments. In this process, data is used to analyze a range of factors such as distance to the hospital, transportation avenues or local weather. Armed with this information generated from AutoML, nurses and support staff have the ability to reach out to the highest-risk patients in more tailored ways with gentle reminders and comforting reassurances. As a result, ARH decreased its cancellation and no-show rates, leading to millions of dollars in savings, and increased the health outcomes of its patients by giving them the care they need, when they need it.

The Journey Starts with Your Data

In the realms of both traditional and generative AI, there is a multitude of untapped potential. Traditional AI is centered on detecting patterns, generating insights, automation and predictions that have been propelling businesses forward for years. Generative AI is in its infancy and starts with a prompt that allows a user to submit a question along with any relevant data to guide content generation without rigorous data preparation or processes to develop. Finding new ways to leverage traditional AI and ML alongside generative AI will drive incremental organizational value from data. And it all starts with a proven commodity – your data.

The post Under Generative AI’s Shadow, Traditional AI Delivers Impact Now appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise technology. In this feature, Qlik‘s Chief Strategy Officer James Fisher offers commentary on making generative AI work for your business via these four key practices.

As we explored in my first article on generative AI, there is an incredible push and pull going on in business today related to the hype of this new technology. On the one hand, organizations are being bombarded with new information about generative AI and many understand that it could provide a stark, first-mover competitive advantage. On the other hand, adoption of new technology is difficult and leaders are cautious about going too fast due to risk, governance and trust concerns.

While we are definitely in a hype cycle with generative AI, it is not the only kind of AI that is beneficial and worth deploying. In fact, other forms of AI have been around much longer and are already an integral part of many data and analytics solutions. As McKinsey notes, “other applications of AI…continue to account for the majority of the overall potential value of AI.” My company, Qlik, first introduced AI into our products five years ago with natural language processing and generation. Today, with our holistic set of AI solutions, Qlik Staige, customers can innovate and move faster by making secure and governed AI part of everything they can do with our technology. This includes experimenting with and implementing generative AI models to develop AI-powered predictions. So, while generative AI is today’s hot topic, we shouldn’t lose sight of the fact that other forms of AI will continue to be very valuable, and if anything, new generative AI tools can actually complement traditional forms of the technology that you’re likely already using in some form.

This brings me back to the ultimate question I hear from business leaders all the time: With the influx of AI-powered solutions on the market, is it even possible to cut through all the noise and drive value for your business with AI? My answer is always a resounding yes. To start, below, I’ll outline three great applications where you can benefit from combining generative AI with current data and analytics strategies.

Making Generative AI Work for Your Business

Ask Real-Time Questions, Get Real-Time Answers

The ability to ask questions and get answers in real-time is a very notable use case for integrating generative AI into your data and analytics efforts. The ideal scenario would be business users asking any question they want and getting contextually relevant content, with the most up-to-date responses available, from AI. Even better is combining this with small subsets of data from an analytics platform in real-time to greatly enrich the context and value of a business’ internal analytics. For example, a customer service representative could generate relevant information about the selections they make based on their real-time interaction with the customer. This arms a business with additional and more effective ways to support its customers, driving home a “customer-first” mindset, which is proven to drive more business value. PwC underlined the importance of this in one of its customer-intelligence deep dives, noting that a shocking 46 percent of all consumers will abandon a brand if its employees do not seem knowledgeable.

Implement Sentiment Analysis to Supercharge Customer Service

Another strong generative AI integration use case is sentiment analysis – a process to determine if a piece of text, like a sentence or a social media post, expresses positive, negative or neutral feelings. In best-case scenarios, businesses would be able to enrich text-based data sets, like product reviews, surveys or service tickets, by using AI to generate sentiment analysis. For example, once a service ticket is identified as “positive” or “negative,” the customer service representative would be able to formulate an appropriate response. Organizations that take it one step further could also implement automation to formulate generative AI-suggested responses to the negative tickets, then feed those suggested responses directly back into a CRM, service database or an analytics application for the customer service representative or other end-user to utilize. This equips organizations with an easier way to diffuse and resolve customer issues more seamlessly, always good for the bottom line.

Use Natural Language to Drive High-Value Insights

Pending the capabilities of a businesses’ analytics platform, one of the best ways I’ve seen to integrate generative AI is through augmenting analytics with natural language and incorporating third-party data into existing data models. In this case, business users would be able to ask specific questions in natural language of their data and receive answers back from their platform. A step further would allow enterprises to also augment existing data and KPIs with a narrative summary. For example, audio electronics company Harman uses ChatGPT on top of its analytics platform to use natural language to drive high value insights with the Qlik analytics engine. This enables analysts to run queries completely on chat searches and receive high-value answers and new prompts for additional analysis.

 The Sky is the Limit

Generative AI is already enabling businesses by bolstering analytics use across the enterprise and helping data experts be more efficient in their work. It spreads data literacy more seamlessly and encourages even novices to dig in. Ultimately, at this stage, new use cases for generative AI in data and analytics are discovered every day. AI has long held the potential to transform business: and generative AI is certainly helping to move the needle forward.

The post Making Generative AI Work for Your Business appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .

Solutions Review’s Contributed Content Series is a collection of contributed articles written by thought leaders in enterprise technology. In this feature, Qlik‘s Chief Strategy Officer James Fisher offers commentary on why the generative AI journey begins with your data and the steps to start.

The daily deluge of generative AI headlines, new vendors, and tech enhancements are enough to evoke anxiety in even the most steadfast executives. Many ponder whether to hop on the generative AI bandwagon quickly, which can sometimes result in missteps, or ignore the hype and potentially get left behind in a highly competitive race. What we found, unsurprisingly, in our own generative AI benchmark report is that 44 percent of enterprises have invested in the technology without any sort of strategy. While I cannot necessarily take away these conflicting feelings, I can assure business leaders that focusing on getting your data house in order will set your organization up to harness the potential of innovative technologies like generative AI – now and in the future.

Fundamentally, generative AI, like all AI, is all about data. Without high-quality and relevant data, generative AI can easily return results that are incomplete or false. Using any old data will not do. I fully agree with McKinsey, who notes that a modern data fabric is a key component for a successful approach to generative AI. This requires organizations to commit to an unwavering process of accessing high-quality, well-harmonized data supported by a scalable data architecture with proper governance and security measures in place.

Harnessing AI’s power will require several actions, such as deploying AI for advanced use cases with a trusted solution. As a start, below are four core areas to fine-tune when it comes to your data and any generative AI or traditional AI effort.

The Generative AI Journey Begins with Your Data

Establish Strong Data Governance 

Businesses are already required to comply with numerous data security and compliance regulations and processes. Poor data quality or uncontrolled data can compromise an enterprise’s decision-making, digital experiences, and operational efficiency and even stifle innovation. These issues are compounded once you add AI into the mix as leaders become concerned with how to maintain the security, privacy, and governance of an organization’s data and how to mitigate the risk of false conclusions based on inaccurate or incomplete information. Data must be organized and trusted to be used as a reliable source for both core operations and AI, so it is imperative to find a data quality solution that meets the needs of the business.

Ensure First-Rate Data Variety

The ability to bring data together from multiple sources – including, but not limited to, SaaS apps, mainframe, files, SAP – from multiple locations and in various formats in real-time is essential to using traditional and generative AI. It fuels the effectiveness and efficiency of AI applications by giving access to a diverse range of information, facilitating more complete and precise analysis. AI models will then be able to unearth more valuable insights, recognize deeper patterns and serve up more informed predictions. Synthesizing data from different sources also strengthens the overall quality and robustness of AI algorithms by lessening biases and other limitations that may arise from relying on a single dataset. This can be done by finding a data integration partner that can bridge connections and deliver information in real-time.

Strive for Connected Systems 

Generative AI will have a truly significant impact on the business when its outputs can be driven directly into operational systems through analytics to truly automate processes that trigger action for business units. Integrated AI-powered analytics can continuously analyze incoming data in real-time, providing immediate insights or recommendations based on pre-defined triggers or thresholds. These triggers can be created with the help of generative AI and then be programmed to launch precise actions or alerts when certain conditions are met. Imagine creating an AI model that analyzes customer data in real-time and detects a potential churn risk. It can trigger a pre-defined automated email campaign almost immediately to re-engage the customer. This is why a no-code automation solution can be the linchpin that generates sophisticated workflows, connected to all business applications and triggers action from each system.

Generate Easily Consumable Insights 

AI systems can generate large amounts of data and insights on their own, but it is critical for businesses to serve up these insights in a format that humans can easily comprehend, collaborate around in real-time and directly drive action from. Digestible formats span both text and visible representations such as graphs, charts or interactive visualizations within the context of other analytics. Representing the outputs of AI in a user-friendly format drives data literacy, making it easier for more people at multiple levels of the enterprise to engage with applications and rely on them for solving complex business issues. Whatever the analytics solution is for a business, it needs to marry dynamic visualizations and dashboards with AI and ML to generate the most meaningful insights that drive value.

Take a Data-First Approach with Generative AI

It is a transformative time in business and technology. Accenture underlines that “solving the data challenge is an urgent priority for every business,” making the race to leverage generative AI and new technologies one about data first. Prepare your business for the exciting future ahead and build your data house with a rock-solid foundation.

In my next piece, I’ll cut through the noise and outline some potential use cases for generative AI integration within analytics.

The post The Generative AI Journey Begins with Your Data appeared first on Best Business Intelligence and Data Analytics Tools, Software, Solutions & Vendors .