In recent years, artificial intelligence (AI) has become more available and accessible than ever before, making it a game-changing technology across various industries. One of the primary reasons behind this surge in popularity is the emergence of generic AI systems, which are off-the-shelf AI solutions trained on general data. These solutions are not tailored specifically to your business or customer needs, but their widespread adoption and rapid growth, as exemplified by platforms such as ChatGPT and Dall-E, have generated quite a buzz. These platforms managed to attract millions of users within just months of their launch, sparking conversations and debates around their potential applications and limitations.

However, amidst this hype, it’s important to remember that many businesses still rely on in-house AI systems that utilize structured tabular data, which may seem relatively mundane compared to the cutting-edge capabilities of generic AI. This begs the question: which approach will truly give you the competitive edge in today’s rapidly evolving market – generic AI or in-house tabular data AI?

There are pros and cons to both approaches, and the following insights can help you make informed decisions about which AI strategy is best suited for your business.

In spite of its ubiquitous presence in pop culture, AI remains a concept without a universally accepted definition. Here, AI is defined as a computer system capable of performing tasks that historically required human intelligence. This definition is helpful as it is not limited to specific types of algorithms, such as neural networks, but differentiates AI from common computing use cases like spreadsheets and databases.

The buzz surrounding AI stems from its potential to revolutionize industries and act as the catalyst for the “fourth industrial revolution.” The latest generation of generative AI systems can create images and text that are convincing enough to be mistaken for human-generated content. As visual learners, humans are wired to process sights and sounds quickly, making it no surprise that text-to-image generators and platforms like ChatGPT have gone viral.

Generic AI: A Good Fit in the Right Situations

Generic AI systems make sense in certain cases, such as when dealing with generic skills like voice recognition or language translation. Some AI-powered tasks require massive amounts of data and computational power, which may be too resource-intensive for all but the largest enterprises to build. However, off-the-shelf AI systems also mean that your competitors have access to the same AI tools, leading to the same decisions, words, and images. Generic AI systems don’t account for your unique customers, products, services, data, business rules, or expert employees.

Tabular Data: Giving Organizations a Competitive Edge

On the other hand, tabular data can offer a competitive edge. Businesses and organizations typically lock this data behind firewalls, making it inaccessible. Your competitors cannot benefit from your proprietary data, and generic AI systems did not have access to this proprietary data when they were trained.

Maybe that’s why more than half of the AI industry experts who participated in a recent survey voted that internal tabular data will give organizations the biggest competitive advantage.

Challenges, and Unlocked Potential (and Value) in Generic AI

Despite its potential, tabular data has not delivered as much value as expected. While it is easier to work with when kept simple, real-life data is often more complex and challenging to use. Simple and easy data manipulation does not create much value; instead, it is the intricacies of real-life data that hold the key to unlocking its potential. The best AI solutions use more than raw data from your database, supplementing learning with inputs such as business goals, domain knowledge, and feature engineering for effective AI applications.

Real-life data typically presents several challenges that make it more difficult to use, including one-to-many relationships across tables in a database, missing or incorrect values, and spanning time with potential structural changes. Additionally, it requires context and domain knowledge, feature engineering to become AI-ready, and is sourced from an almost infinite set of schemas without well-defined AI-specific semantics. Furthermore, it changes quite frequently, adding another layer of complexity.

As a result, generating value from tabular data requires significantly more effort, but overcoming these challenges can unlock its true potential in AI-driven solutions.

For Competitive Differentiation: Focus on Data Quality and Domain Knowledge

To make tabular data competitive, it is crucial to focus on data quality, as poor data leads to poor outcomes. Feature engineering is necessary to carefully select and transform data for successful model building. Domain knowledge is also essential to ensure that the algorithm learns the correct business rules and addresses the right business problems. This deep understanding of underlying data allows businesses to harness the power of their proprietary information and leverage AI to gain a competitive advantage.

In today’s rapidly evolving business landscape, it is essential to consider whether you want to be behind, equal to, or ahead of your competitors when it comes to adopting AI technologies. By not leveraging AI systems like ChatGPT, you risk falling behind as your competitors embrace these advancements. However, simply using generic AI solutions may only put you on equal footing with others in your industry.

To truly gain a competitive edge, it is vital to harness your unique intellectual property by building AI systems that are trained on your proprietary data and business rules, and incorporating insights from your valuable subject matter experts. By capitalizing on your organization’s distinctive strengths and knowledge, you can elevate your AI strategy beyond that of your competitors, ensuring that your business remains at the forefront of innovation.

[To share your insights with us, please write to sghosh@martechseries.com]

The post Generic AI or Tabular Data AI: Which One Gives You the Competitive Edge? appeared first on AiThority.

The AI research team at the University of Warwick has developed an intelligent machine learning-based algorithm as part of social media monitoring and cluster intelligence. This AI-based social media intelligence algorithm can be used to identify and evaluate how people on social media communicate about their opinion, experiences and concerns toward vaccinations. This model is called “The Vaccine Attitude Detection (VADet) Model.” The latest addition to our coverage of AI and machine learning projects exemplify how advanced data science could be used effectively in improving our interaction with machines and internet.

VADeT model is an advanced ML dataset that requires minimal training. It can be trained using a small sample dataset of online tweets before these are used for larger analyses. The new ML model can identify how people’s attitudes vary in the online domain and how different kinds of fact-checking and conspiracy tool kits work in the social media world as far as vaccinations are concerned. AI researchers at the University of Warwick believe their AI model could potentially save healthcare organizations and government agencies save millions of dollars wasted in creating awareness drives about vaccination. By leveraging social media platforms, healthcare institutions can channelize their resources to address vaccine misinformation and negative comments posted on social media platforms such as Facebook, Instagram, LinkedIn, Twitter, and TikTok.

The AI-based model can analyse a social media post and establish its author’s stance towards vaccines, by being ‘trained’ to recognise that stance from a small number of example tweets. – University of Warwick

The UK Research and Innovation (UKRI) funded this research led by Professor Yulan He. It was proposed to be presented at the Annual Conference of the North American Chapter of the Association for Computational Linguistics on 12 July 2022.

What is VADeT?

The COVID-19 pandemic is the most destructive event of our lifetime. Vaccinations have helped save many lives. Yet, online social media platforms are filled with negative sentiments and comments linked to vaccinations– which could emerge from a variety of sources such as fear, anxiety and misinformation. Many organizations collected data from social media platforms such as Twitter to understand how healthcare companies could use Search and Big Data strategy to improve rate of vaccinations.

Recommended: Meet DailyTalk: The Latest Conversational Text-to-Speech Dataset based on FastSpeech Framework

VADeT is a powerful AI-based algorithm running on semi-supervised approach for sentiment analysis and tweet clustering on topics related to vaccinations. It is majorly focused at simplifying the process linked to annotated and unlabelled data. It uses a variational auto-encoding architecture to learn from a very small dataset of unlabelled data and then augments the analysis for fine-tuning annotated data of user attitudes and sentiment extracted from social media posts. Currently, it is proposed for use as a tool to combat “infodemic” but in the long run, it could also be used against fake news detection and marketing analysis.

Benefits of Using AI ML: AI Adoption Could Save Businesses $35,000 Every Year

Currently, it has been proved that VADet is able to learn disentangled stance and aspect topics. It also outperforms existing aspect-based sentiment analysis models on both stance detection and tweet clustering.

AI Machine Learning Projects on Vaccination and How People Generally Talk about it Online

Public sentiment has always been divided about vaccinations, much before social media came into the picture. However, online access to social media micro-blogging sites has certainly played a big role in spreading misinformation about vaccines. But, thanks to recent developments in Natural Language Processing, Deep Learning and Sentiment Analysis, AI researchers are able to exactly identify the extent to which social media comments and conversations influence vaccinations success in recent regions of the world. From vaccine hesitancy to anti-government stance, AI is able to find what really put people off about vaccinations.

The recent AI research by the University of Warwick follows the recent demand for social media intelligence to curb infodemic about vaccinations and their potential benefits and effectiveness, especially in the Post-COVID-19 era. For example, AI and machine learning researchers at the Stevens Institute of Technology were on their way toward developing a scalable solution: an AI tool capable of detecting “fake news” relating to COVID-19. It gathered data from 24000 tweets to develop a stance detection algorithm.

While there are many challenges in the way AI and machine learning are used to detect and analyze vaccination misinformation, there role in creating more awareness about the benefits of vaccination can’t be underestimated.

Top AI ML Insights: DeepNash and the World of Model-free Multi-agent Reinforcement Learning (RL)

[To share your insights with us, please write to sghosh@martechseries.com]

The post AI Researchers Use Social Media Monitoring Tactics to Identify Behavior Toward Vaccination appeared first on AiThority.

At the core of every AI algorithm are three basic ingredients:

1) the ability to measure,

2) knowing how much of what you measure needs to be processed, and, of course,

3) the ability to process more than one input at a time.

To what depth a system can measure can be thought of as its potential. Determining what aspects of those measurements must be sent to the processor can be thought of as delivering that potential. Finally, knowing how to combine the salient parts of those measurements in the correct proportions, known as sensor fusion, is the key to exploring an algorithm’s IQ or reasoning potential. Augment that sensor fusion with a feedback loop and the algorithm will have the ability to check and course-correct its logic, a necessary ingredient in machine learning.  

These three attributes are the key to understanding the depth of an AI’s unique power.

And like many things, the more you cultivate and calibrate these foundational elements, the better the AI algorithm will perform in the long term. Now that we understand the three areas to explore let’s dive into the first component, measurement depth, and how it’s critical to the foundation of building a robust, high-performing AI algorithm.  

Measurement depth 

Metrology is the study of measurement science and measurement depth plays a crucial role in building a robust algorithm. The Gagemaker’s Rule, or 10:1 rule, states that a measurement device must be 10x more precise than the desired measurement. The reason that measurement depth is so critical is that it determines the possible level of precision and sets the algorithm’s maximum potential. Therefore, the more precision you have in any given measurement, the greater the AI algorithm’s potential.

Top AI Machine Learning Insights: AI Platform Tethr Announces Zero Touch Conversation Intelligence

Metrology focuses on the deep understanding of a particular measurement. That measurement can be as simple and distinct as voltage, ground, or temperature or as multi-modal as the functioning of aircraft control surfaces, or as complex as maximizing throughput on a manufacturing assembly line. Whether you are measuring a single parameter or several, the depth of each measurement determines the level of programmability that’s possible. For instance, measuring a 3 Volt system to 1/10th of a volt is not as insightful as measuring to 1/1000th of a volt. Depending on the system that voltage is powering, the extra precision may be critical for battery life or maybe a distraction. Maximizing the potential of any algorithm requires matching the entire end-to-end measurement needs to the needed depth. This is true no matter what’s being measured, even data systems, which may not be as immediately intuitive, so let’s look at one of those examples.  

How to optimize measurement 

Enterprise IT stacks are now a complex web of interconnected data systems, each exchanging information aimed at tuning an organization’s operations. These technology stacks include an array of software such as CRM, ERP, databases, order fulfillment, and each with unique data formats and custom application programming interfaces (APIs). According to Salesforce, the average company has over  900 applications in its tech stack, many of them cloud-based and all of them experiencing software updates that can have ripple impacts. Identifying and isolating problems, much less optimizing performance across multiple intersecting software applications, is akin to finding a needle in a collection of interconnected haystacks. 

Recommended AI ML News: InRule Technology Democratizes Machine Learning by Enhancing Platform with AutoML Capabilities

Each software application in a tech stack has a different sponsor in an enterprise – finance, human resources, sales, marketing, supply chain – and that primary org’s considerations are top of mind for IT. Every enterprise has custom workflows and integrations with numerous applications and backend systems, and user journeys span various paths and are rarely linear. Therefore, even if two enterprises used identical applications in their tech stack, mapping all the exchange points and validating the end-to-end operation would be unique. If there were ever an application in need of AI, this would be it. The measurements, in this case, could be the intersystem data input points, the intrasystem data exchange points, and the data display points. 

Understanding how an AI algorithm would operate in a system like this would start with understanding how it measures points data in three key areas:  

1. Measuring how users interface with the application, regardless of the operating system, which in some cases involves employing robotic process automation (RPA) when button pushes are required
2. Measuring the data exchanges between and command APIs that link the systems in a complex technology stack to ensure they are occurring correctly
3. Measuring on-screen information across omni platforms (desktops and mobile) such as images, text and logos as a human would to see how they render

Evaluating the measurement efficacy starts with its ability to measure regardless of the operating system, software versions, devices, or interface mechanisms. The more conditions under which the AI algorithms cannot measure, the less impactful it will be in operation.  

Conclusion

Whenever you assess the potential of anything, start with the foundation.

At the foundation of every AI system is its ability to measure. The more it can measure, the more impactful it has the potential of being. Look at what it is capable of measuring and, more importantly, where it is not capable. Limited sensing results in limited AI algorithm potential. The old adage from Lord Kelvin stands true today that “if you cannot measure it, you cannot improve it.” To understand the true power of any AI, make sure to start by analyzing its measurement breadth and depth.

[To share your insights with us, please write to sghosh@martechseries.com]

The post Determining the Potential of Your AI Algorithm Starts with Measurement appeared first on AiThority.

As the field of applied AI matures and there are more and more capabilities available to be used, it is often less a question of how to build powerful technology and more one of how to achieve a meaningful impact with that technology.

At the end of the day, AI is a tool after all.

It is a means to an end, not an end in it of itself. In the technology space, we care about the successful adoption and deployment of AI because of the innumerable problems it can solve. Yet, time and again we see companies, of all shapes and sizes, blindly invest in or deploy AI without a clear understanding of what problems they are looking to solve or the value they are trying to generate. This results in huge amounts of wasted time and money.

But it doesn’t have to be this way.

Regardless of the type of AI or the problem you are dealing with, there are a few key areas that are worth considering in order to deploy AI effectively:

• Problem Identification and Definition
• Evaluation and Expectation
• Experimentation and Improvement

Each one of these areas has its own challenges but ignoring any of them can jeopardize whole projects or initiatives. Here is how companies and their technology teams can best approach each of these areas and drive AI adoption success.

Problem Identification and Definition

The main challenge in applied Data Science today is not how to solve a problem but knowing what problem to solve to begin with. Clearly scoping and understanding a problem is often the hardest challenge we face as technologists and business leaders. While certain problems will require varying degrees of technological complexity and innovation to solve, without an initial hypothesis it will be incredibly difficult to create impact. We often see isolated teams of data scientists given a vague mandate who may go on to create technically impressive prototypes or Proof of Concepts (POCs.) Their work may even result in academic publications, but what they build fails to be deployed within the business and create any user value or impact.

Deep Learning Anomaly Detection Gets Big Boost with Imagimob AI’s tinyML Platform

There are a number of reasons for this but solving a problem that is not a priority or one that does not create value is often part of the story.

Recommended AI ML Blog; Biggest AI-based Telehealth Updates for 2022

Evaluation and Expectation

After problem definition, one of the most common challenges in applied AI is a misalignment in expectations of performance among stakeholders. What is good enough? What type of mistakes are acceptable, and which are unacceptable? What role does subjectivity and potential or perceived bias in data sets play?

How important are trust and explainability?

These are just a few questions where the agreement needs to be reached and where expectations may initially be misaligned. And this misalignment isn’t just limited to between internal and external teams but is commonplace among teams “in-house.”

For instance, when dealing with sarcasm detection, a product manager might have an expectation that anything with less than 70% accuracy will not meet a customer’s expectation while a data scientist might see 60% accuracy as a massive achievement given the complexity of the problem. Without alignment, this can lead to tension and debate over whether an AI solution is “good enough” to be deployed to customers. The risk is that you invest in building an AI solution that is then determined to not be accurate enough to be used by anyone. It is imperative to align on expectations first before trying to solve a problem instead of trying to do so retroactively.

Once these expectations have been agreed upon, it is important that stakeholders settle on the proper metrics and methods of evaluation. This is obviously easier said than done, however, to start with it is key that business leaders and technology teams keep two prevailing facts in mind when building evaluation frameworks: all AI systems will eventually make a mistake; and even though AI will often outperform humans, it will sometimes make really stupid mistakes that a human would never make (e.g., an automatic camera following the bald head of a referee instead of the ball during a soccer match). Mistakes such as these can be discouraging and cause decision-makers to completely scrap AI projects but they do not necessarily indicate that an AI solution is ineffective.

By keeping these two things in mind, businesses can approach evaluating AI solutions more realistically.

Experimenting and Improving in AI Adoption Stages

Once we have a clear problem to solve with aligned expectations of performance around agreed evaluation metrics, we can start experimenting to find the best possible solution to the problem. This has been, arguably, the step with the most attention in the data science and broader technology community over the years.

The well-known, and well-ignored advice is to start with the simplest possible solution that might work end to end. Once this solution is built, and assuming there is a clear evaluation framework to compare different alternatives, we can experiment and compare as many approaches as we want and select the best one given the needs. Here we need to remember to be problem-driven. Sometimes an off-the-shelf solution will do the trick and engineers with machine learning expertise can integrate such solutions into a broader offering.

Other times you may need novel research and will want to involve data scientists with deeper expertise in a given problem space. But it is important to not just assume that novel research is needed without examining the powerful existing solutions that already exist.

When experimenting it is also important to remember that effective AI cannot be created in a vacuum. Without the right technological readiness, it will be challenging to build AI in the first place and then impossible to deploy and maintain it thereafter.

Machine Learning Operations or MLOps has adapted much of traditional DevOps to the data science life cycle. Importing concepts like continuous improvement and development as well as ongoing monitoring. Such an environment is necessary to create ease of experimentation with minimum friction for testing new models and ideas. If your data scientists are able to build POCs and prototypes easily, these can be shown either to clients or internally earlier in order to iterate more quickly.

After this initial experimentation when a system is in production and integrated into your product it is important to remember that AI solutions are not “fire and forget.”

Once live, they need to be monitored and potentially re-tuned or even changed completely over time. Again, this sounds obvious to anyone who has worked in software development. However, due to the relative immaturity of AI, we often do not see widespread adoption of MLOps practices like continual improvement and monitoring.

Editorial note: Miguel Martinez, Co-Founder and Chief Data Scientist at Signal AI is a co-author of this post

[To share your insights with us, please write to sghosh@martechseries.com]

The post Applied AI Adoption: How to Get it Right appeared first on AiThority.

Tata Consultancy Services is Launching 15 Phygital Learning Programs Available Across India to Skill and Upskill the Youth of the Nation

TCS iON™, a strategic unit of Tata Consultancy Services (TCS) and Nettur Technical Training Foundation (NTTF), a premier technical and vocational education and training institute, have come together to launch high-demand skill development programs across sectors, in the areas of robotics, automation, manufacturing, and electronics, in a unique phygital model, developed by the former.

TCS and NTTF will offer 3 diploma and 12 certification courses with a target to skill and upskill more than 60,000 youth in the country, making them job-ready for current and future industry needs. Designed in line with the recent academic reforms and industry-defined standards, these learning programs, launched by TCS iON, aim to bridge the skills gap. Enrolled candidates will have the flexibility to learn anytime, anywhere, under the mentorship of best-in-class trainers from the industry and NTTF, and receive hands-on experience at the TCS iON learning and practice centers set up across the country.

Download Our Top Whitepaper : Building Reliable and Secure Fintech Systems in 2022

TCS iON’s phygital model transforms the learning ecosystem by combining hands-on project work and multimodal digital learning resources. The live online lectures will be delivered by NTTF. These programs will be offered to existing students as well as alumni of various ITI, polytechnics and skill development institutes across the country who have partnered with TCS.

Venguswamy Ramaswamy, Global Head, TCS iON, said, “The Indian electronics manufacturing service industry is expected to grow over six times from $23.5 billion to reach $152 billion by 2025. To realize this growth, the country requires its youth to acquire vocational skills in addition to core education. The TCS iON-NTTF partnership is aligned with the National Education Policy 2020, aimed to make the future workforce skilled and equipped to pursue exciting career opportunities in emerging industries.”

Dr N Reguraj, Managing Director, NTTF, said, “The race for the skilled youth is now global. The Phygital model programs being launched as a collaborative initiative by NTTF and TCS iON would bridge the huge skill gap among the youth in the country. Our target is to skill and upskill over 60,000 youth to make them job-ready. This unique program offered as a blended model of hands-on training plus the online content delivery as per NTTF standards would make the youth “Multi-level Certified” and skillful, to suit the high-demand job roles across sectors like robotics, automation, Industry 4.0, and electronics. The successful students will have an opportunity to choose among innumerable job postings listed on the TCS iON platform.”

Skill development institutes, ITIs and polytechnic colleges across India can offer these programs to interested learners. Apart from the newly launched courses, TCS iON is also set to launch several other courses which are designed to transform India into a skilled workforce for achieving the goal of becoming Atmanirbhar Bharat.

Recommended AI News: Clarify Health Acquires Embedded Healthcare to Scale Value-Based Care for Health Plans

[To share your insights with us, please write to sghosh@martechseries.com]

The post TCS iON Partners with NTTF to Launch Industry-Led Skilling Programs appeared first on AiThority.

Chatbots are dynamic agents with the express capability to engage in conversational interactions. By applying innovative linguistic reduction rules to user utterances, we empower chatbots to reduce any statement or sentence into its most basic form so bots can swiftly understand it and appropriately respond.

The relationship between linguistic reduction rules and chatbots for natural language technology applications is two-fold. First, this pairing drastically simplifies chatbot applications so that no matter what text or speech the chatbot encounters, they can readily understand and respond to it. Secondly, by adding elements of knowledge graphs and taxonomies to this tandem, the resulting combination can make chatbots more useful than any current commercial offerings —including Alexa and Siri.

Reductions Simplify Language

The general concept behind this symbolic reasoning approach is that when people speak or write they use more words than necessary to produce the simplest logical statement they’re conveying. For example, there are numerous ways to ask someone his or her name, including “Could you tell me your name, please?”

Reduction rules would reduce this simple question to “What is your name?”, so bots can quickly comprehend its meaning, then use additional techniques to answer it.

Although this example seems trivial, it illustrates the basic formula that’s integral for revamping a host of business use cases from analyzing legal documents to forms for regulatory compliance and heightening call center interactions—or any other NLP application.

Recommended: Klocked Announces Siri Shortcuts Feature

Less Enables More 

Linguistic reductions can be considered a rule-based approach, one of the foundations of the symbolic Artificial Intelligence approach to NLP. Although they’re manually created, their beauty lies in their universal applicability to any natural language processing use case. This utility significantly broadens when taxonomies are involved, but even without such hierarchical vocabularies, this approach works in any domain – from pharmaceutical to finance or any other domain. These rules are based on identifying patterns in language and reducing it to its bare minimum so chatbots or computers can easily understand them.

Consider the following sentence a customer might ask: ‘I just need to know if you’re open at the present time.’ Because reduction rules are recursive, bots can apply a series of them to this question.

For instance, one might state that anytime ‘I just’ is detected, it can be reduced to ‘I’. Or, phrases beginning with ‘I need to know something’ are reducible to ‘tell me something’. By applying these and other rules, this 12-word sentence (complicated by a contraction, adverbs, a prepositional phrase, and other superfluities) is reduced to ‘are you open now?”—a much simpler sentence.

NLP Technologies Update: Arthur Releases the First NLP Model Monitoring Solution To Serve Soaring Enterprise Adoption

Reduction Rules

Think of reduction rules as containing a left side—the input or all the exhaustive ways something can be expressed in language—and a right side or output containing the reduction in its simplest form.

Chatbots combine these rules with an additional set of rules for responses to questions, which is useful for online interactions with a retailer, vendor, or certain websites. However, the response rules could easily be replaced with rules about data privacy regulations, additional regulations, or legal implications for clauses in contracts and other documents.

Generating these rules requires training data or several labeled examples of common questions, phrases, and sentences found in natural language itself. However, the benefit is definite quality control in the bots’ responses, which are suitable for each of the specifically-reduced sentences. These controlled outputs are ideal for ensuring legal teams about the accuracy and respectfulness of chatbot interactions—which isn’t always guaranteed when only machine learning techniques are involved. Outputs may also contain a multitude of responses that mean the same thing for a given reduction, so bots have vibrant personalities.

Building Chatbot Intelligence 

The real magic of doing NLP with chatbots and reduction rules comes with combining them with domain-specific knowledge graphs powered by taxonomies. In this case, the questions customers ask, or the information found in legal documents, would serve as a query to run against a knowledge base to devise timely answers. The query itself would be stored in the knowledge graph (which relies on a uniform taxonomy) to constantly build knowledge to support highly sophisticated use cases, including anecdotal applications.

Whereas most chatbots breakdown after an initial sentence, storing each of the reductions in an episode about how a dishwasher stopped working, for example, as triples in a knowledge graph allows firms to query them for holistic understanding of everything in that story. Applying this technique to maintenance records of aircraft repairs, for example, could deliver insight into common repair scenarios and predictive maintenance to eliminate failure altogether.

By enhancing chatbots with linguistic reduction rules and knowledge graphs, it is possible to give chatbots truly conversational sophistication exceeding anything existent today.

About the Authors

Jans Aasman is a Ph.D. psychologist, expert in Cognitive Science and CEO of Franz Inc., an early innovator in Artificial Intelligence and leading provider of Semantic Database technology and Knowledge Graph solutions.

Dr. Richard Wallace is a knowledge engineer with over 35 years of experience in data science and artificial intelligence. He is a co-founder of Pandorabots, Inc., a leading chatbot A.I. company. In the 2000’s Dr. Wallace was a three-time winner of the Loebner Prize in artificial intelligence for the “most human computer.”

[To share your insights with us, please write to sghosh@martechseries.com]

The post Linguistic Reduction and Knowledge Graphs for Next-Gen Chatbots appeared first on AiThority.

AI ethics and empathy have become the key disruptors in the age of hard-core AI and Machine learning deployment. Organizations are finding it extremely hard to balance their AI roadmap around governance, ethics, and empathy. AI bias has been one of the biggest problem areas in the data science industry that none has managed to solve. Yet, we are optimistic that things will change in the coming years. Why do I think so? Some of the biggest technology companies are taking it upon themselves to remove AI biases and encourage activities that strengthen AI ethics in the industry. Pega is one such company.

In 2020, Pega become one of the first companies to bring out a technology-driven platform that helps eliminate biases hidden in the artificial intelligence (AI) driving customer engagements. The feature flags possible discriminatory offers and messages generated by AI across all channels before they reach the customer. Despite so much happening in AI ethics, we still find biases and loopholes in the way AI is deployed across departments at an organization level. In order to better understand the scope of Ethical applications of AI in the current scenario, we chatted with Vince Jeffs, Senior Director – Marketing, AI and Decisioning at Pega.

Here’s our predictions series featuring Vince Jeffs and his ideas on AI ethics and how ethical applications would influence business operations in 2022.

Hi, Vince, tell us about the most fascinating thing about working in AI and machine learning domain?

Ethics in AI.

Why do you feel ethics in AI is so exciting?

It’s been a fashionable trope in recent years for both businesses and governments to talk about using AI in a way that is both ethical and responsible.

I think it’s high time that Ethical/Responsible AI moves beyond ‘fluffy policy’ and becomes embedded inside the tangible tools and actual law and regulations.

What’s in store for AI for enterprise applications? How much should companies invest in AI technologies?

AI will become established as a business tool.

Traditionally, AI has been the preserve of a select few, many of whom have expert qualifications and work in fields like data science where the technology can allow them to recognize patterns in large subsets of data. Next year, this could change, and we’ll see the use of AI moving beyond the data scientists and statisticians and into the realm of data-savvy business people who will be able to use to drive better agility, understand their customers more and drive better outcomes.

How do you see businesses taking a lead and making amends to their policies when it comes to AI deployment?

Organizations have become well-versed and even better practiced at telling us how important this is, yet, actual steps to ensure their impact has been much rarer.

In 2022, that should change, and AI will move into the realm of solid regulation and law being put into practice.

For example, the EU has put out a new proposal, which next year it will be looking to move forward as the first steps towards getting beyond rhetoric and into regulation. This legislation will prohibit the use of AI with unacceptable risk to do harm, such as, with some narrow exceptions, the use of AI for biometric identification (such as facial recognition) in public spaces for law enforcement, the use of subliminal techniques to distort behavior so that personal or physical harm is caused,,and using AI for social scoring by public authorities (a situation where social media commentary can significantly impact on your ability to get a job, vote, or get credit).

How is AI regulated in the current context of commercial applications?

Applications of AI with high risk will become closely regulated. The UK has also just published its National AI strategy, and the Chinese government in August announced proposed regulation that in certain areas will go further than the EU. Similar developments in the USA will also take things a step further, we could see organizations finding themselves having to prove that they are not just complying with regulations around ethics and responsibility in the way they are using AI, but also that they are using it to benefit customers and provide them with the transparency and explainability required to reassure consumers that it is being used as a force for good.

Could you highlight the good things that would happen in AI with better control?

Certainly! Better operationalization of AI will make ‘evil’ apps a thing of the past.

As more responsible AI becomes a greater priority for the majority of businesses, attention will turn to eliminate ‘evil’ apps, in which the AI has been deliberately manipulated to provide nefarious outcomes, such as ransomware, where software has been deliberately designed to block access to a computer system until a sum of money is paid. Alternatively, other ‘evil’ apps can arise as a result of problems in the algorithm which lead to unintended consequences, such as automated decisioning systems that include some form of undesired algorithmic bias toward protected groups. Both instances can be overcome by providing greater rigor and structure around how these apps are developed and why. By answering and documenting simple questions, the use of ‘evil apps’ can be reduced. These questions include: What information does the app contain and why?

How does it work? What are you building and why?

What is this app’s purpose?

Can you prove that algorithmic bias is within limits, and can automated decisions explain themselves?

We could see ultimate transparency appear within the app development space where a full register of what developers are doing with the AI is a prerequisite to producing anything.

We hear a lot about making AI more human. Could you provide more context to this imaginative thought and how it could become a reality in the future?

The rise of ‘The Intelligence of Everything’ will help AI become more ‘human’.

This year we’ll see AI beginning to become more well-rounded and go beyond the intellectual functions we associate it mainly with today, and we’ll begin to see it evolve into emotional, creative and relational intelligence and other more abstract ‘human’ qualities. We are not just intellectual beings – it’s these other qualities that make us truly human. By replicating and/or exploring the human condition through the AI lens, businesses will be able to better understand their customers’ emotional state, interact and bond more naturally and emphatically in chatbots and other channels, and provide a better overall service.

What does the future look like for AI companies? What surprises are in store for us with AI becoming more mainstream?

2022 will be the year people finally go ‘all in’ on AI.

It’s fair to say that AI has had a tricky infancy, childhood and adolescence. There’s no doubt its role has changed considerably – from its initial introduction on the edge of a business in their innovation labs to the present day when people are beginning to understand that it has the capacity to transform organizations from the center out. In recent years there’s been caution about extending its use beyond basic functionality, and how much it can be trusted, which has meant its use hasn’t been pervasive within businesses. However, now that more and more organizations have dipped their toe into the water and have had their eyes opened as to the benefits it can provide, the technology is finally ready to reach maturity. A key reason for that is that end users are also reaching maturity in their own understanding about both how they can get the best results from AI, and also the rights and wrongs of using it. Now that AI has been largely demystified, users have a far better understanding of how to apply it effectively and correctly, which means that they are finally ready to adopt it on a wider basis and send its use into the mainstream. AI is leaving the labs and transforming the business from the core.

Enterprises’ lofty top-level goals to become more evidence-centric and data-driven are translated into pervasive, and ubiquitous automated decisions that drive and optimize any customer interaction or business process.

Thank you Vince for chatting with us and sharing your exemplary ideas about putting Ethics at the center of AI deployment. We look forward to speaking with you again.

The post Predictions Series 2022: Everything You Want to Know about AI Ethics appeared first on AiThority.

Microsoft Synapse ML is now live. Microsoft has renamed and rebranded its open-source library for scalable machine learning pipelines MMLSpark to SynapseML to attract the open-source DevOps community. The new platform would empower developers to extract more out of scalable ML pipelines and unify each of these into a simplified ecosystem that works well with a variety of AI ML programming languages such as Python, R, Scala, and Java, and others.

Recommended AI ML Project Updates: McCombs School of Business Hosts Global Summit on Explainable AI

Why SynapseML

Machine learning engineers are on the lookout for tailor-made production-ready ML models. A few years ago, thinking of readymade ML models would have been a crazy proposition. But today, open-source ML libraries are flooding the industry with these solutions, suitably developed with “glue” codes for different ecosystems.

With SynapseML, developers can build scalable and intelligent systems for solving challenges in domains such as:

Top NFT News: Reason Launches World’s First NFT Escape Room

Microsoft has released a graphical layout to explain how SynapseML aligns with modern ML development pipelines.

The ML frameworks would further refine industry solutions in Azure Synapse Analytics, making it easy for users to leverage custom-build ML models for a variety of applications, including for e-commerce and retail, healthcare, customer support, telecom, and so on. Microsoft SynapseML would integrate with existing ML systems offered by Microsoft such as Azure Cognitive Services, Light GBM, Vowpal Wabbit, MLFLow, AzureML and other Spark workflows.

[To share your insights with us, please write to sghosh@martechseries.com]

The post Microsoft SynapseML Launched: A Rebranded Version of Open-source ML Library MMLSpark appeared first on AiThority.

The Center for Analytics and Transformative Technologies at The University of Texas McCombs School of Business will host its flagship annual conference online Nov. 11-12 to discuss the future of Explainable AI. The global AI summit would specifically focus on the most contentious topics related to cracking open the “black box” of artificial intelligence using next-gen data analytics and human intelligence.

This year’s CATT Global Analytics Summit, organized by McCombs School of Business, invites experts from industry and academia to explore the theme, “Explainable AI: Building Trust and Transparency in Data Systems and Machine Learning Models.”

“In many organizations, AI and machine learning tools are becoming more powerful and sophisticated, but the problem is that this level of sophistication can lead to black boxes,” said Michael Sury, CATT managing director and member of McCombs’ finance faculty.

“There can be an opacity to AI processes, and that opacity may not be comfortable for organizational leaders, corporate stakeholders, regulators or customers,” he said.

So, the task of explaining how AI arrives at its decisions — including whether those decisions are trustworthy, reliable, and free of bias — has grown increasingly important for business executives and data scientists alike, Sury said. “If the black box is a barrier to adoption of these powerful techniques, how do we overcome that barrier?”

Helping to answer questions such as these are summit keynote speakers Charles Elkan, professor of computer science at the University of California, San Diego and former global head of machine learning at Goldman Sachs in New York; and Cynthia Rudin, professor of computer science at Duke University and principal investigator at Duke’s Interpretable Machine Lab.

The summit will also feature talks and panel discussions moderated by McCombs faculty members. Among the nearly two dozen speakers will be Scott Lundberg, senior researcher at Microsoft, on his trademark “SHAP values” as tools for interpreting AI outputs. Krishnaram Kenthapadi, principal scientist at Amazon, will address responsible AI in the industry.

The topics are broadly applicable across organizations, Sury said: “The implications of AI explainability are sweeping, including in areas like risk management, ethics, compliance, reliability and customer relationship management.”

Last year’s summit, also online, drew more than 1,700 registrants from 50 countries. Registration is free again this year, and registrants will receive a program booklet after the event.

[To share your insights with us, please write to sghosh@martechseries.com]

The post McCombs School of Business Hosts Global Summit on Explainable AI appeared first on AiThority.

Businesses are in a tearing hurry to become AI-fueled organizations. However, only a handful of these actually manages to truly become one among many in the race.  In a recent report on what makes AI-fueled organizations successful, Deloitte AI Institute has uncovered startling facts related to the association between strong corporate leadership, AI strategy executions, and financial budgeting in AI and machine learning projects.

For example, an enterprise-wide AI strategy and leadership that communicate a bold vision are nearly 1.7 times more likely to achieve high outcomes.

Strategy Leading Practices That Help You Succeed as an AI-Fueled Organization

AI-fueled organizations view AI as a key element of business differentiation and success, and they set an enterprise-wide strategy that is championed from the top.

• Transformers were three times more likely to have an enterprise-wide strategy in place, and twice as likely as Starters to report a differentiated AI approach.
• However, only 40% of survey respondents completely agreed that their company has an enterprise-wide AI strategy in place, leaving room for improvement.
• While 66% of respondents view AI as critical to success, only 38% believe their use of AI differentiates them from competitors.

Deloitte AI Institute’s fourth edition of the “State of AI in the Enterprise” survey, conducted between March and May 2021, explores the deeper transformations happening inside organizations that are using AI to drive value in order to understand what the most “AI-fueled organizations” are doing to drive success. The report finds that AI-fueled organizations leverage data as an asset to deploy and scale AI systematically across all types of core business processes in a human-centered way.

Top Salesforce Interviews: AiThority Interview with Julian Armington, Senior Director Product Marketing at Salesforce

At the time of this announcement, Nitin Mittal, Deloitte AI co-leader, and principal, Deloitte Consulting LLP said – “Becoming an AI-fueled organization is to understand that the transformation process is never complete, but rather a journey of continuous learning and improvement.”

By surveying 2,875 executives from 11 countries across the Americas, Europe, and Asia, the report identifies the leading practices that were aligned with the marketplace leaders across industries. The findings in the report aim to help companies overcome challenges to becoming an AI-fueled organization, no matter what stage of AI transformation they are in, and especially to help those who are earlier in their transformation.

Profiles in AI maturity

In order to assess organizations’ AI maturity, Deloitte grouped responding organizations into four profiles based on how many types of AI applications they have deployed full-scale and the number of outcomes achieved to a high degree.

• Twenty-eight percent (28%) of survey respondents are “Transformers,” who report high outcomes and a high number of AI deployments. This group has identified and largely adopted leading practices associated with the strongest AI outcomes.
• Twenty-six percent (26%) of respondents are “Pathseekers,” reporting high outcomes, but a low number of deployments. Pathseekers have adopted capabilities and behaviors that are leading to success, but on fewer initiatives and they have not scaled to the same degree as Transformers.
• “Underachievers” were 17% of respondents, reporting low outcomes and a high number of deployments. While these organizations have a significant amount of AI deployment activity, they haven’t adopted enough leading practices to help them effectively achieve meaningful outcomes.
• “Starters” — 29% of survey respondents — reported low outcomes and a low number of deployments. These organizations have gotten a late start in building AI capabilities and are the least likely to demonstrate leading practice behaviors.

Recommended: AiThority Interview with Dinesh Nirmal, General Manager at IBM Automation

“By embracing AI strategically and challenging orthodoxies, organizations can define a roadmap for adoption, quality delivery and scale to create or unlock value faster than ever before.”
— Irfan Saif, Deloitte AI co-leader, and principal, Deloitte & Touche LLP

“The risks associated with AI remain top of mind for executives. We found that high-achieving organizations report being more prepared to manage risks associated with AI and confident that they can deploy AI initiatives in a trustworthy way.”
— Beena Ammanath, executive director of the Deloitte AI Institute, Deloitte Consulting LLP.

Leading practices drive AI success

In analyzing the four organizational profiles, the report identified the behaviors most associated with strong outcomes, and leading practices in the following categories: strategy, operations, culture and change management, and ecosystems.

Operations leading practice

AI-fueled organizations establish new operating models and processes that drive sustained quality, innovation, and value creation.

• Organizations that have undergone significant changes to workflows or added new roles are almost 1.5 times more likely to achieve outcomes to a high degree.
• Organizations that document and enforce MLOps processes are twice as likely to achieve their goals; nearly two times more likely to report being extremely prepared for risks associated with AI; and nearly two times more confident that they can deploy AI initiatives in a trustworthy way.
• Survey results exposed a disconnect, however, as only about one-third of respondents report that they have adopted leading operational practices for AI.

Culture and Change Management Leading Practice Boost AI ML Adoption

AI-fueled organizations nurture a trusting, agile, data-fluent culture and invest in change management to support new ways of working.

• Organizations that invest in change management to a high degree are 1.6 times more likely to report that AI initiatives exceed expectations and over 1.5 times more likely to achieve their desired goals.
• However, most organizations underinvest this area. Only 37% of survey respondents reported significant investment in change management, incentives, or training activities to help their people integrate new technology into their work, often resulting in a slower, less successful transformation.

Ecosystems leading practice

AI-fueled organizations orchestrate dynamic ecosystems that build and protect competitive differentiation.

• Eighty-three percent (83%) of the highest achieving organizations (Transformers and Pathfinders) create a diverse ecosystem of partnerships to execute their AI strategy.
• Organizations with diverse ecosystems are significantly more likely to have a transformative vision for AI, enterprise-wide AI strategies, and use AI as a strategic differentiator.

The Deloitte AI Institute supports the positive growth and development of AI through engaged conversations and innovative research. It also focuses on building ecosystem relationships that help advance human-machine collaboration in the “Age of With,” a world where humans work side-by-side with machines.

[To share your insights with us, please write to sghosh@martechseries.com]

The post Deloitte AI Institute Uncovers Key Driving Forces in an AI-fueled Organization appeared first on AiThority.