A Closer Look at Generative AI

Artificial intelligence is already designing microchips and sending us spam, so what's next? Here's how generative AI really works and what to expect now that it's here.

Generative AI is an umbrella term for any kind of automated process that uses algorithms to produce, manipulate, or synthesize data, often in the form of images or human-readable text. It's called generative because the AI creates something that didn't previously exist. That's what makes it different from discriminative AI, which draws distinctions between different kinds of input. To say it differently, discriminative AI tries to answer a question like "Is this image a drawing of a rabbit or a lion?" whereas generative AI responds to prompts like "Draw me a picture of a lion and a rabbit sitting next to each other."

This article introduces you to generative AI and its uses with popular models like ChatGPT and DALL-E. We'll also consider the limitations of the technology, including why "too many fingers" has become a dead giveaway for artificially generated art.

The emergence of generative AI

Generative AI has been around for years, arguably since ELIZA, a chatbot that simulates talking to a therapist, was developed at MIT in 1966. But years of work on AI and machine learning have recently come to fruition with the release of new generative AI systems. You've almost certainly heard about ChatGPT, a text-based AI chatbot that produces remarkably human-like prose. DALL-E and Stable Diffusion have also drawn attention for their ability to create vibrant and realistic images based on text prompts. We often refer to these systems and others like them as models because they represent an attempt to simulate or model some aspect of the real world based on a subset (sometimes a very large one) of information about it.

Output from these systems is so uncanny that it has many people asking philosophical questions about the nature of consciousness—and worrying about the economic impact of generative AI on human jobs. But while all these artificial intelligence creations are undeniably big news, there is arguably less going on beneath the surface than some may assume. We'll get to some of those big-picture questions in a moment. First, let's look at what's going on under the hood of models like ChatGPT and DALL-E.

How does generative AI work?

Generative AI uses machine learning to process a huge amount of visual or textual data, much of which is scraped from the internet, and then determine what things are most likely to appear near other things. Much of the programming work of generative AI goes into creating algorithms that can distinguish the "things" of interest to the AI's creators—words and sentences in the case of chatbots like ChatGPT, or visual elements for DALL-E. But fundamentally, generative AI creates its output by assessing an enormous corpus of data on which it’s been trained, then responding to prompts with something that falls within the realm of probability as determined by that corpus.

Autocomplete—when your cell phone or Gmail suggests what the remainder of the word or sentence you're typing might be—is a low-level form of generative AI. Models like ChatGPT and DALL-E just take the idea to significantly more advanced heights.

Training generative AI models

The process by which models are developed to accommodate all this data is called training. A couple of underlying techniques are at play here for different types of models. ChatGPT uses what's called a transformer (that's what the T stands for). A transformer derives meaning from long sequences of text to understand how different words or semantic components might be related to one another, then determine how likely they are to occur in proximity to one another. These transformers are run unsupervised on a vast corpus of natural language text in a process called pretraining (that's the P in ChatGPT), before being fine-tuned by human beings interacting with the model.

Another technique used to train models is what's known as a generative adversarial network, or GAN. In this technique, you have two algorithms competing against one another. One is generating text or images based on probabilities derived from a big data set; the other is a discriminative AI, which has been trained by humans to assess whether that output is real or AI-generated. The generative AI repeatedly tries to "trick" the discriminative AI, automatically adapting to favor outcomes that are successful. Once the generative AI consistently "wins" this competition, the discriminative AI gets fine-tuned by humans and the process begins anew.

One of the most important things to keep in mind here is that, while there is human intervention in the training process, most of the learning and adapting happens automatically. So many iterations are required to get the models to the point where they produce interesting results that automation is essential. The process is quite computationally intensive. 

Is generative AI sentient?

The mathematics and coding that go into creating and training generative AI models are quite complex, and well beyond the scope of this article. But if you interact with the models that are the end result of this process, the experience can be decidedly uncanny. You can get DALL-E to produce things that look like real works of art. You can have conversations with ChatGPT that feel like a conversation with another human. Have researchers truly created a thinking machine?

Chris Phipps, a former IBM natural language processing lead who worked on Watson AI products, says no. He describes ChatGPT as a "very good prediction machine." Phipps says: 'It’s very good at predicting what humans will find coherent. It’s not always coherent (it mostly is) but that’s not because ChatGPT "understands." It’s the opposite: humans who consume the output are really good at making any implicit assumption we need in order to make the output make sense.'

Phipps, who's also a comedy performer, draws a comparison to a common improv game called Mind Meld: 'Two people each think of a word, then say it aloud simultaneously—you might say "boot" and I say "tree." We came up with those words completely independently and at first, they had nothing to do with each other. The next two participants take those two words and try to come up with something they have in common and say that aloud at the same time. The game continues until two participants say the same word.

Maybe two people both say "lumberjack." It seems like magic, but really it’s that we use our human brains to reason about the input ("boot" and "tree") and find a connection. We do the work of understanding, not the machine. There’s a lot more of that going on with ChatGPT and DALL-E than people are admitting. ChatGPT can write a story, but we humans do a lot of work to make it make sense.'

Testing the limits of computer intelligence

Certain prompts that we can give to these AI models will make Phipps' point fairly evident. For instance, consider the riddle "What weighs more, a pound of lead or a pound of feathers?" The answer, of course, is that they weigh the same (one pound), even though our instinct or common sense might tell us that the feathers are lighter.

ChatGPT will answer this riddle correctly, and you might assume it does so because it is a coldly logical computer that doesn't have any "common sense" to trip it up. But that's not what's going on under the hood. ChatGPT isn't logically reasoning out the answer; it's just generating output based on its predictions of what should follow a question about a pound of feathers and a pound of lead. Since its training set includes a bunch of text explaining the riddle, it assembles a version of that correct answer. But if you ask ChatGPT whether two pounds of feathers are heavier than a pound of lead, it will confidently tell you they weigh the same amount, because that's still the most likely output to a prompt about feathers and lead, based on its training set. It can be fun to tell the AI that it's wrong and watch it flounder in response; I got it to apologize to me for its mistake and then suggest that two pounds of feathers weigh four times as much as a pound of lead.

Why does AI art have too many fingers?

A notable quirk of AI art is that it often represents people with profoundly weird hands. The "weird hands quirk" is becoming a common indicator that the art was artificially generated. This oddity offers more insight into how generative AI does (and doesn't) work. Start with the corpus that DALL-E and similar visual generative AI tools are pulling from: pictures of people usually provide a good look at their face but their hands are often partially obscured or shown at odd angles, so you can't see all the fingers at once. Add to that the fact that hands are structurally complex—they're notoriously difficult for people, even trained artists, to draw. And one thing that DALL-E isn't doing is assembling an elaborate 3D model of hands based on the various 2D depictions in its training set. That's not how it works. DALL-E doesn't even necessarily know that "hands" is a coherent category of thing to be reasoned about. All it can do is try to predict, based on the images it has, what a similar image might look like. Despite huge amounts of training data, those predictions often fall short.

Phipps speculates that one factor is a lack of negative input: 'It mostly trains on positive examples, as far as I know. They didn't give it a picture of a seven fingered hand and tell it "NO! Bad example of a hand. Don’t do this." So it predicts the space of the possible, not the space of the impossible. Basically, it was never told to not create a seven fingered hand.'

There's also the factor that these models don't think of the drawings they're making as a coherent whole; rather, they assemble a series of components that are likely to be in proximity to one another, as shown by the training data. DALL-E may not know that a hand is supposed to have five fingers, but it does know that a finger is likely to be immediately adjacent to another finger. So, sometimes, it just keeps adding fingers. (You can get the same results with teeth.) In fact, even this description of DALL-E's process is probably anthropomorphizing it too much; as Phipps says, "I doubt it has even the understanding of a finger. More likely, it is predicting pixel color, and finger-colored pixels tend to be next to other finger-colored pixels."

Potential negative impacts of generative AI

These examples show you one of the major limitations of generative AI: what those in the industry call hallucinations, which is a perhaps misleading term for output that is, by the standards of humans who use it, false or incorrect. All computer systems occasionally produce mistakes, of course, but these errors are particularly problematic because end users are unlikely to spot them easily: If you are asking a production AI chatbot a question, you generally won't know the answer yourself. You are also more likely to accept an answer delivered in the confident, fully idiomatic prose that ChatGPT and other models like it produce, even if the information is incorrect. 

Even if a generative AI could produce output that's hallucination-free, there are various potential negative impacts:

• Cheap and easy content creation: Hopefully it's clear by now that ChatGPT and other generative AIs are not real minds capable of creative output or insight. But the truth is that not everything that's written or drawn needs to be particularly creative. Many research papers at the high school or college undergraduate level only aim to synthesize publicly available data, which makes them a perfect target for generative AI. And the fact that synthetic prose or art can now be produced automatically, at a superhuman scale, may have weird or unforeseen results. Spam artists are already using ChatGPT to write phishing emails, for instance.
• Intellectual property: Who owns an AI-generated image or text? If a copyrighted work forms part of an AI's training set, is the AI "plagiarizing" that work when it generates synthetic data, even if it doesn't copy it word for word? These are thorny, untested legal questions.
• Bias: The content produced by generative AI is entirely determined by the underlying data on which it's trained. Because that data is produced by humans with all their flaws and biases, the generated results can also be flawed and biased, especially if they operate without human guardrails. OpenAI, the company that created ChatGPT, put safeguards in the model before opening it to public use that prevent it from doing things like using racial slurs; however, others have claimed that these sorts of safety measures represent their own kind of bias.
• Power consumption: In addition to heady philosophical questions, generative AI raises some very practical issues: for one thing, training a generative AI model is hugely compute intensive. This can result in big cloud computing bills for companies trying to get into this space, and ultimately raises the question of whether the increased power consumption—and, ultimately, greenhouse gas emissions—is worth the final result. (We also see this question come up regarding cryptocurrencies and blockchain technology.)

Use cases for generative AI

Despite these potential problems, the promise of generative AI is hard to miss. ChatGPT's ability to extract useful information from huge data sets in response to natural language queries has search giants salivating. Microsoft is testing its own AI chatbot, dubbed "Sydney," though it's still in beta and the results have been decidedly mixed.

But Phipps thinks that more specialized types of search are a perfect fit for this technology. "One of my last customers at IBM was a large international shipping company that also had a billion-dollar supply chain consulting side business," he says.

Phipps adds: 'Their problem was that they couldn’t hire and train entry level supply chain consultants fast enough—they were losing out on business because they couldn’t get simple customer questions answered quickly. We built a chatbot to help entry level consultants search the company's extensive library of supply chain manuals and presentations that they could turn around to the customer.If I were to build a solution for that same customer today, just a year after I built the first one, I would 100% use ChatGPT and it would likely be far superior to the one I built. What’s nice about that use case is that there is still an expert human-in-the-loop double-checking the answer. That mitigates a lot of the ethical issues. There is a huge market for those kinds of intelligent search tools meant for experts.'

Other potential use cases include:

• Code generation: The idea that generative AI might write computer code for us has been bubbling around for years now. It turns out that large language models like ChatGPT can understand programming languages as well as natural spoken languages, and while generative AI probably isn't going to replace programmers in the immediate future, it can help increase their productivity.
• Cheap and easy content creation: As much as this one is a concern (listed above), it's also an opportunity. The same AI that writes spam emails can write legitimate marketing emails, and there's been an explosion of AI copywriting startups. Generative AI thrives when it comes to highly structured forms of prose that don't require much creativity, like resumes and cover letters.
• Engineering design: Visual art and natural language have gotten a lot of attention in the generative AI space because they're easy for ordinary people to grasp. But similar techniques are being used to design everything from microchips to new drugs—and will almost certainly enter the IT architecture design space soon enough.

Conclusion

Generative AI will surely disrupt some industries and will alter—or eliminate—many jobs. Articles like this one will continue to be written by human beings, however, at least for now. CNET recently tried putting generative AI to work writing articles but the effort foundered on a wave of hallucinations. If you're worried, you may want to get in on the hot new job of tomorrow: AI prompt engineering.

Author: Josh Fruhlinger

Source: InfoWorld 

Chatbots and their Struggle with Negation

Today’s language models are more sophisticated than ever, but they still struggle with the concept of negation. That’s unlikely to change anytime soon.

Nora Kassner suspected her computer wasn’t as smart as people thought. In October 2018, Google released a language model algorithm called BERT, which Kassner, a researcher in the same field, quickly loaded on her laptop. It was Google’s first language model that was self-taught on a massive volume of online data. Like her peers, Kassner was impressed that BERT could complete users’ sentences and answer simple questions. It seemed as if the large language model (LLM) could read text like a human (or better).

But Kassner, at the time a graduate student at Ludwig Maximilian University of Munich, remained skeptical. She felt LLMs should understand what their answers mean — and what they don’t mean. It’s one thing to know that a bird can fly. “A model should automatically also know that the negated statement — ‘a bird cannot fly’ — is false,” she said. But when she and her adviser, Hinrich Schütze, tested BERT and two other LLMs in 2019, they found that the models behaved as if words like “not” were invisible.

Since then, LLMs have skyrocketed in size and ability. “The algorithm itself is still similar to what we had before. But the scale and the performance is really astonishing,” said Ding Zhao, who leads the Safe Artificial Intelligence Lab at Carnegie Mellon University.

But while chatbots have improved their humanlike performances, they still have trouble with negation. They know what it means if a bird can’t fly, but they collapse when confronted with more complicated logic involving words like “not,” which is trivial to a human.

“Large language models work better than any system we have ever had before,” said Pascale Fung, an AI researcher at the Hong Kong University of Science and Technology. “Why do they struggle with something that’s seemingly simple while it’s demonstrating amazing power in other things that we don’t expect it to?” Recent studies have finally started to explain the difficulties, and what programmers can do to get around them. But researchers still don’t understand whether machines will ever truly know the word “no.”

Making Connections

It’s hard to coax a computer into reading and writing like a human. Machines excel at storing lots of data and blasting through complex calculations, so developers build LLMs as neural networks: statistical models that assess how objects (words, in this case) relate to one another. Each linguistic relationship carries some weight, and that weight — fine-tuned during training — codifies the relationship’s strength. For example, “rat” relates more to “rodent” than “pizza,” even if some rats have been known to enjoy a good slice.

In the same way that your smartphone’s keyboard learns that you follow “good” with “morning,” LLMs sequentially predict the next word in a block of text. The bigger the data set used to train them, the better the predictions, and as the amount of data used to train the models has increased enormously, dozens of emergent behaviors have bubbled up. Chatbots have learned style, syntax and tone, for example, all on their own. “An early problem was that they completely could not detect emotional language at all. And now they can,” said Kathleen Carley, a computer scientist at Carnegie Mellon. Carley uses LLMs for “sentiment analysis,” which is all about extracting emotional language from large data sets — an approach used for things like mining social media for opinions.

So new models should get the right answers more reliably. “But we’re not applying reasoning,” Carley said. “We’re just applying a kind of mathematical change.” And, unsurprisingly, experts are finding gaps where these models diverge from how humans read.

No Negatives

Unlike humans, LLMs process language by turning it into math. This helps them excel at generating text — by predicting likely combinations of text — but it comes at a cost.

“The problem is that the task of prediction is not equivalent to the task of understanding,” said Allyson Ettinger, a computational linguist at the University of Chicago. Like Kassner, Ettinger tests how language models fare on tasks that seem easy to humans. In 2019, for example, Ettinger tested BERT with diagnostics pulled from experiments designed to test human language ability. The model’s abilities weren’t consistent. For example:

He caught the pass and scored another touchdown. There was nothing he enjoyed more than a good game of ____. (BERT correctly predicted “football.”)

The snow had piled up on the drive so high that they couldn’t get the car out. When Albert woke up, his father handed him a ____. (BERT incorrectly guessed “note,” “letter,” “gun.”)

And when it came to negation, BERT consistently struggled.

A robin is not a ____. (BERT predicted “robin,” and “bird.”)

On the one hand, it’s a reasonable mistake. “In very many contexts, ‘robin’ and ‘bird’ are going to be predictive of one another because they’re probably going to co-occur very frequently,” Ettinger said. On the other hand, any human can see it’s wrong.

By 2023, OpenAI’s ChatGPT and Google’s bot, Bard, had improved enough to predict that Albert’s father had handed him a shovel instead of a gun. Again, this was likely the result of increased and improved data, which allowed for better mathematical predictions.

But the concept of negation still tripped up the chatbots. Consider the prompt, “What animals don’t have paws or lay eggs, but have wings?” Bard replied, “No animals.” ChatGPT correctly replied bats, but also included flying squirrels and flying lemurs, which do not have wings. In general, “negation [failures] tended to be fairly consistent as models got larger,” Ettinger said. “General world knowledge doesn’t help.”

Invisible Words

The obvious question becomes: Why don’t the phrases “do not” or “is not” simply prompt the machine to ignore the best predictions from “do” and “is”?

That failure is not an accident. Negations like “not,” “never” and “none” are known as stop words, which are functional rather than descriptive. Compare them to words like “bird” and “rat” that have clear meanings. Stop words, in contrast, don’t add content on their own. Other examples include “a,” “the” and “with.”

“Some models filter out stop words to increase the efficiency,” said Izunna Okpala, a doctoral candidate at the University of Cincinnati who works on perception analysis. Nixing every “a” and so on makes it easier to analyze a text’s descriptive content. You don’t lose meaning by dropping every “the.” But the process sweeps out negations as well, meaning most LLMs just ignore them.

So why can’t LLMs just learn what stop words mean? Ultimately, because “meaning” is something orthogonal to how these models work. Negations matter to us because we’re equipped to grasp what those words do. But models learn “meaning” from mathematical weights: “Rose” appears often with “flower,” “red” with “smell.” And it’s impossible to learn what “not” is this way.

Kassner says the training data is also to blame, and more of it won’t necessarily solve the problem. Models mainly train on affirmative sentences because that’s how people communicate most effectively. “If I say I’m born on a certain date, that automatically excludes all the other dates,” Kassner said. “I wouldn’t say ‘I’m not born on that date.’”

This dearth of negative statements undermines a model’s training. “It’s harder for models to generate factually correct negated sentences, because the models haven’t seen that many,” Kassner said.

Untangling the Not

If more training data isn’t the solution, what might work? Clues come from an analysis posted to arxiv.org in March, where Myeongjun Jang and Thomas Lukasiewicz, computer scientists at the University of Oxford (Lukasiewicz is also at the Vienna University of Technology), tested ChatGPT’s negation skills. They found that ChatGPT was a little better at negation than earlier LLMs, even though the way LLMs learned remained unchanged. “It is quite a surprising result,” Jang said. He believes the secret weapon was human feedback.

The ChatGPT algorithm had been fine-tuned with “human-in-the-loop” learning, where people validate responses and suggest improvements. So when users noticed ChatGPT floundering with simple negation, they reported that poor performance, allowing the algorithm to eventually get it right.

John Schulman, a developer of ChatGPT, described in a recent lecture how human feedback was also key to another improvement: getting ChatGPT to respond “I don’t know” when confused by a prompt, such as one involving negation. “Being able to abstain from answering is very important,” Kassner said. Sometimes “I don’t know” is the answer.

Yet even this approach leaves gaps. When Kassner prompted ChatGPT with “Alice is not born in Germany. Is Alice born in Hamburg?” the bot still replied that it didn’t know. She also noticed it fumbling with double negatives like “Alice does not know that she does not know the painter of the Mona Lisa.”

“It’s not a problem that is naturally solved by the way that learning works in language models,” Lukasiewicz said. “So the important thing is to find ways to solve that.”

One option is to add an extra layer of language processing to negation. Okpala developed one such algorithm for sentiment analysis. His team’s paper, posted on arxiv.org in February, describes applying a library called WordHoard to catch and capture negation words like “not” and antonyms in general. It’s a simple algorithm that researchers can plug into their own tools and language models. “It proves to have higher accuracy compared to just doing sentiment analysis alone,” Okpala said. When he combined his code and WordHoard with three common sentiment analyzers, they all improved in accuracy in extracting opinions — the best one by 35%.

Another option is to modify the training data. When working with BERT, Kassner used texts with an equal number of affirmative and negated statements. The approach helped boost performance in simple cases where antonyms (“bad”) could replace negations (“not good”). But this is not a perfect fix, since “not good” doesn’t always mean “bad.” The space of “what’s not” is simply too big for machines to sift through. “It’s not interpretable,” Fung said. “You’re not me. You’re not shoes. You’re not an infinite amount of things.” 

Finally, since LLMs have surprised us with their abilities before, it’s possible even larger models with even more training will eventually learn to handle negation on their own. Jang and Lukasiewicz are hopeful that diverse training data, beyond just words, will help. “Language is not only described by text alone,” Lukasiewicz said. “Language describes anything. Vision, audio.” OpenAI’s new GPT-4 integrates text, audio and visuals, making it reportedly the largest “multimodal” LLM to date.

Future Not Clear

But while these techniques, together with greater processing and data, might lead to chatbots that can master negation, most researchers remain skeptical. “We can’t actually guarantee that that will happen,” Ettinger said. She suspects it’ll require a fundamental shift, moving language models away from their current objective of predicting words.

After all, when children learn language, they’re not attempting to predict words, they’re just mapping words to concepts. They’re “making judgments like ‘is this true’ or ‘is this not true’ about the world,” Ettinger said.

If an LLM could separate true from false this way, it would open the possibilities dramatically. “The negation problem might go away when the LLM models have a closer resemblance to humans,” Okpala said.

Of course, this might just be switching one problem for another. “We need better theories of how humans recognize meaning and how people interpret texts,” Carley said. “There’s just a lot less money put into understanding how people think than there is to making better algorithms.”

And dissecting how LLMs fail is getting harder, too. State-of-the-art models aren’t as transparent as they used to be, so researchers evaluate them based on inputs and outputs, rather than on what happens in the middle. “It’s just proxy,” Fung said. “It’s not a theoretical proof.” So what progress we have seen isn’t even well understood.

And Kassner suspects that the rate of improvement will slow in the future. “I would have never imagined the breakthroughs and the gains we’ve seen in such a short amount of time,” she said. “I was always quite skeptical whether just scaling models and putting more and more data in it is enough. And I would still argue it’s not.”

Date: June 2, 2023

Author: Max G. Levy

Source: Quanta Magazine

From Patterns to Predictions: Harnessing the Potential of Data Mining in Business

Data mining techniques can be applied across various business domains such as operations, finance, sales, marketing, and supply chain management, among others. When executed effectively, data mining provides a trove of valuable information, empowering you to gain a competitive advantage through enhanced strategic decision-making.

From Visualization to Analytics: Generative AI's Data Mastery

Believe it or not, generative AI is more than just text in a box. The truth is that it transcends the boundaries of traditional creative applications. So what it does is it extends the capabilities of the user far beyond text generation. It’s an art. In addition to its prowess in crafting captivating narratives and artistic creations, generative AI demonstrates its versatility by helping users empower their own data analytics. 

With its advanced algorithms and language comprehension, it can navigate complex datasets and distill valuable insights. This transformative shift underscores the convergence of creativity and analysis, as generative AI empowers users to harness its intelligence for data-driven decision-making. 

From uncovering hidden patterns to providing actionable recommendations, generative AI’s proficiency in data analytics heralds a new era where innovation spans the spectrum from artistic expression to informed business strategies. 

So let’s take a brief look at some examples of how generative AI can be used for data analytics. 

Datasets for Analysis

Our first example is its capacity to perform data analysis when provided with a dataset. Imagine equipping generative AI with a dataset rich in information from various sources. Through its proficient understanding of language and patterns, it can swiftly navigate and comprehend the data, extracting meaningful insights that might have remained hidden by the casual viewer. Even experts can miss patterns after a while, but for AI, it’s made to detect them.

All of this goes beyond mere computation. By crafting human-readable summaries and explanations, AI is able to make the findings accessible to a wider audience, especially to non-expert stakeholders who may not have a deep-level understanding of what they’re being shown. 

This symbiotic fusion of data analysis and natural language generation underscores AI’s role as a versatile partner in unraveling the layers of information that drive informed decisions.

Data Visualization Through Charts

The second example of how generative AI is multifaceted is its ability to create user-friendly charts that seamlessly integrate with other data visualization tools. Suppose you have a dataset and require a visual representation that’s both insightful and easily transferable to other programs. Generative AI can step up to the plate by creating charts that are not only visually appealing but also tailored to your data’s characteristics. 

Whether it’s a bar graph, scatter plot, or line chart, generative AI can provide charts ready for your preferred mode of visualization. This streamlined process bridges the gap between data analysis and visualization, empowering users to effortlessly harness their data’s potential for impactful presentations and strategic insights.

Idea Generation

This isn’t isolated to just data analytics. Most marketers have found that generative AI tools are great at this. That’s because the technology is great at helping its human users with idea generation and refining concepts by acting as a collaborative brainstorming partner. Consider a scenario where you’re exploring a new project or problem-solving endeavor. Engaging generative AI allows you to bounce ideas off of it, unveiling a host of potential questions and perspectives that might not have otherwise occurred to you. 

Through its adept analysis of the input and context, generative AI not only generates thought-provoking questions but also offers insights that help you delve deeper into your topic. This relationship between the human user and the AI transforms generative AI into an invaluable ally, driving the exploration of ideas, prompting critical thinking, and guiding the conversation toward uncharted territories of creativity and innovation.

Cleaning Up Data and Finding Anomalies

As mentioned above, generative AI has a knack for finding patterns, and these patterns aren’t just isolated to being positive. With a good generative AI program, a data team can take on even the meticulous task of data cleaning and anomaly detection. Picture a dataset laden with imperfections and anomalies that could skew analysis results. The AI can be harnessed to comb through the data, identifying inconsistencies, outliers, and irregularities that might otherwise go unnoticed. 

Again, AI has a keen eye for patterns and deviations to aid in ensuring the integrity of the dataset. Human error is human error, but with AI, that error can be reduced significantly. Furthermore, generative AI doesn’t just flag anomalies—it provides insights into potential causes and implications. This fusion of data cleaning and analysis empowers users to navigate the complexities of their data landscape with confidence, making informed decisions based on reliable, refined datasets.

Creating Synthetic Data

Synthetic data generation is yet another facet where generative AI’s adaptability shines. When faced with limited or sensitive datasets, the AI can step in to generate synthetic data that mimics the characteristics of the original information. This synthetic data serves as a viable alternative for training models, testing algorithms, and ensuring privacy compliance. By leveraging its understanding of data patterns and structures, 

Generative AI crafts synthetic datasets that maintain statistical fidelity while safeguarding sensitive information. This innovative application showcases generative AI’s role in bridging data gaps and enhancing the robustness of data-driven endeavors, providing a solution that balances the need for accurate analysis with the imperative of data security.

Conclusion

Some great stuff huh? As you have just read, generative AI isn’t only for creating amazing images, or a chatbot that can help office workers with their tasks. It’s a technology that if utilized correctly can help any data professionals supercharge their data analytics. Now, are you ready to learn more?

Date: September 22, 2023

Author:

Source: ODSC

Natural Language Processing is the New Revolution in Healthcare AI

After a late start, Artificial Intelligence (AI) has made massive inroads in penetrating the pharmaceutical and healthcare industries. Nearly all aspects of the industry have felt impacts from the increasing uptake of the technologies – from visual and image diagnostics, to Machine Learning and Neural Networks for drug discovery, data analysis and other models. One of the more recently ascendant forms of AI in healthcare has been Natural Language Processing (NLP) – which promises to revolutionize digitization, communication and analyzing human-generated data. We explore the foundations of the technology and its current and future applications in the industry.

What is Natural Language Processing?

NLP refers to the practice of using computer methods to process language in the form generated by humans – often disordered, flexible and adaptable. The phenomenon is not limited to AI technology; it first originated in the 1950s with symbolic NLP. This was a rudimentary concept which was originally intended for machine translation applications – first demonstrated by the 1954 Georgetown Experiment. Symbolic NLP was largely founded on complex, manually defined rulesets: it soon became apparent that the unpredictability and fluidity of human language could not truly be defined by concise rules. 

With the exponential growth in computing power, symbolic NLP soon gave rise to statistical NLP – largely pioneered by IBM and their Alignment Models for machine translation. IBM developed six such models, which enabled a more flexible approach to machine-based translation. Other companies soon followed, and statistical NLP evolved into what we know today as self-learning NLP, underpinned by machine learning and neural networks. The developments in its ability to recognize and process language have put it to use in fields far more diverse than translation – although it continues to make improvements there too. 

While symbolic NLP is still often employed when datasets are small, statistical NLP has been largely replaced by neural network-enabled NLP. This is due to how neural networks can simplify the process of constructing functional models. The trade-off lies in the opacity of how they operate – while statistical methods will always be fully transparent and the path to how they obtain their results will be fully visible, neural network models are often more of a “black box”. Their power in interpreting human language is not to be underestimated however – from speech recognition, including the smart assistants we have come to rely on, to translations and text analytics, NLP promises to bridge many gaps.

Current Models in Healthcare

One of the most obvious applications of NLP in the healthcare industry is processing written text – whether that be analog or digital. A leading source of data heterogeneity, which often prevents downstream analysis models from directly utilizing datasets, is the different terminology and communication used by healthcare practitioners. Neural-enabled NLPs can condense such unstructured data into directly comparable terms suitable for use in data analysis. This can be seen in models inferring International Classification of Diseases (ICD) codes based on records and medical texts.

Medical records present rich datasets that can be harnessed for a plethora of applications with the appropriate NLP models. Medical text classification using NLPs can assist in classifying disease based on common symptoms – such as identifying the underlying conditions causing obesity in a patient. Models such as this can then be later used to predict disease in patients with similar symptoms. This could prove particularly revolutionary in diseases such as sepsis – which has early symptoms that are very common across a number of conditions. Recurrent Neural Network NLP models using patient records to predict sepsis showed a high accuracy with lower false alarm rates than current benchmarks. 

These implementations are also critical in clinical development. But clinical operations also generate another source of unstructured data: adverse event reports, which form the basis of pharmacovigilance and drug safety. We already explored the applications of AI models in pharmacovigilance in a different article – but their introduction to that field particularly highlights the need for increased cooperation with regulatory authorities to ensure all stakeholders remain in lockstep as we increasingly adopt AI. 

Beyond Healthcare

But NLP can also be applied beyond human language. Exploratory studies have also shown its potential in understanding artificial languages, such as protein coding. Proteins, strings of the same 20 amino acids presenting in variable order, share many similarities with human language. Research has shown that language generation models trained on proteins can be used for many diverse applications, such as predicting protein structures that can evade antibodies. 

There are other sources of unstructured, heterogeneous data that are simply too big to pore over with human eyes in cost-efficient manners. Science literature can be one of these – with countless journal articles floating around libraries and the web. NLP models have previously been employed by companies such as Linguamatics to capture valuable information from throughout the corpus of scientific literature to prioritize drug discovery efforts.

Quantum computing also represents a major growing technology, and firms are already seeking to combine it with NLP. One example is Quantinuum’s λambeq library, which can convert natural language sentences to quantum circuits for use in advanced computing applications. Future endeavors in the area promise massive advancements in text mining, bioinformatics and all the other applications of NLP.

Research conducted by Gradient Flow has shown that Natural Language Processing is the leading trend in AI technologies for healthcare and pharma. This is for good reason – AI can prove useful in a cornucopia of different implementations, but NLP is what facilitates the use of heterogeneous, fragmented datasets in the same model. Integrating existing and historical datasets, or new datasets generated in inherently unstructured manners – articles, records and medical observations, will remain crucial in the progress of other AI technologies. NLP is what enables that – and future advancements are likely to see its prominence rise on its own, as well.

Author: Nick Zoukas

Source: PharmaFeatures

The increasing use of AI-driven chatbots for customer service in Ecommerce

AI-powered chatbots have revolutionized the way Ecommerce businesses handle customer service. With the ability to provide immediate responses and resolutions, chatbots ensure that customers receive prompt assistance at any time of day or night. These chatbots are programmed with natural language processing (NLP) capabilities, allowing them to understand and interpret human language accurately.

Moreover, AI-powered chatbots can collect data on customer interactions and use it for personalization purposes in future exchanges. They can also learn from their previous conversations, continually improving their responses over time. As a result, businesses can offer more informed recommendations and tailored solutions to customers’ problems.

Another significant advantage of AI-powered chatbots is that they help reduce operating costs by eliminating the need for human agents to attend to every customer query. Chatbots can handle multiple queries simultaneously without sacrificing quality or efficiency. This feature allows companies to streamline their operations while still providing an exceptional customer experience.

Benefits of Chatbots

Chatbots have become an essential part of customer service in the Ecommerce industry. One of the significant benefits of using chatbots is their ability to offer 24/7 customer support, which ensures that customers can get assistance at any time they need it. This feature helps businesses reduce wait times and improve customer satisfaction rates.

Another advantage of chatbots in Ecommerce is their efficiency in handling repetitive inquiries. As a result, businesses can free up their staff from handling these inquiries, allowing them to focus on complex tasks that require human intervention. Chatbots also help companies save money by reducing the need for additional staffing during peak periods.

Additionally, chatbots are excellent tools for collecting customer data and providing personalized recommendations based on their purchase history, preferences, and behavior patterns. This allows businesses to provide tailored services to each customer, increasing the likelihood of repeat purchases and improving overall loyalty. The use of AI-powered chatbots also helps companies stay ahead of the competition by offering cutting-edge technology that enhances the overall shopping experience for customers.

Challenges & Risks

One of the challenges that come with using AI-powered chatbots in Ecommerce customer service is ensuring that they are programmed to understand and respond appropriately to all types of customer inquiries. While chatbots have the potential to speed up response times and improve efficiency, they can also risk alienating customers if their responses are generic or irrelevant. As such, a significant amount of resources must be dedicated to developing chatbot algorithms that can handle complex queries and adapt to different situations.

One of the challenges of starting an ecommerce business is related to data privacy and security risks associated with chatbot interactions. Chatbots gather a vast amount of sensitive information from customers, including personal details such as names, addresses, and payment information. This makes them an attractive target for cybercriminals who may try to infiltrate the system and steal this valuable data. Companies must ensure that their security protocols are robust enough to protect against cyber attacks while still providing fast and convenient customer service.

Finally, there is a risk associated with relying too heavily on AI-powered chatbots at the expense of human interaction. While these algorithms can handle many routine tasks effectively, customers may still require personalized attention or assistance for more complex inquiries or issues. Over-reliance on automation may lead to decreased customer satisfaction levels over time as customers demand more direct interaction with human representatives who can provide empathy and context-specific solutions.

Ecommerce Use Cases

One of the most significant use cases for AI-powered chatbots in Ecommerce is customer service. With the ability to handle massive amounts of customer inquiries quickly, chatbots can improve response times and reduce wait times for customers. Additionally, chatbots can offer 24/7 support, which is particularly useful for businesses with global customers who are located in different time zones.

Another key use case for AI-powered chatbots in Ecommerce is product recommendations. By analyzing a customer’s browsing behavior and purchase history, chatbots can offer personalized product recommendations that are tailored to their unique preferences. This not only improves the overall shopping experience but also helps to increase sales by promoting products that customers are more likely to buy.

Finally, AI-powered chatbots can also be used for order tracking and delivery notifications. By providing real-time updates on the status of an order or delivery, chatbots can help reduce anxiety and uncertainty among customers while improving transparency and accountability within the supply chain.

Industry Examples

One industry that has been quick to adopt AI-powered chatbots in their customer service is the Ecommerce industry. With the increased demand for online shopping and a growing number of customers seeking 24/7 support, chatbots have become a valuable tool for Ecommerce businesses to provide efficient and effective customer service. These intelligent virtual assistants can handle multiple queries simultaneously, provide instant responses, and even offer personalized recommendations based on a customer’s purchase history.

Another industry that has leveraged the power of AI-powered chatbots is the banking sector. Banks are using chatbots to enhance their customer service by providing real-time assistance with account inquiries, transaction history, and even fraud detection. In addition to providing prompt responses, some banks have also integrated voice recognition technology into their chatbots to enable customers to complete transactions through voice commands securely.

Overall, AI-powered chatbots have proven to be an innovative solution for various industries looking to streamline their customer service operations. With advances in natural language processing (NLP) and machine learning algorithms, these virtual assistants are continually improving in their ability to understand complex queries and offer personalized solutions – making them an invaluable asset for companies seeking cost-effective ways of delivering exceptional customer experiences.

Future Outlook

The future of customer service in Ecommerce looks promising with the rise of AI-powered chatbots. These chatbots are designed to provide personalized support and assistance to customers, which can significantly improve their overall shopping experience. They are able to handle a variety of tasks, such as answering common questions, providing product recommendations, and even completing purchases.

One major advantage of using chatbots in customer service is that they are available 24/7. Customers no longer have to wait for business hours or deal with long hold times on phone calls. Chatbots can provide quick and efficient support at any time of the day, which can lead to higher customer satisfaction rates.

Looking forward, it’s expected that AI-powered chatbots will continue to evolve and become even more advanced in their capabilities. As they learn from interactions with customers, they will be able to provide increasingly accurate and relevant support. This could ultimately lead to reduced costs for businesses while also improving the overall shopping experience for customers.

Conclusion: Growing Role of AI

In conclusion, the growing role of AI in Ecommerce customer service is becoming increasingly important. The use of chatbots has revolutionized the way businesses interact with their customers. They provide 24/7 support and can handle multiple customer queries at once, leading to faster response times and increased customer satisfaction.

AI-powered chatbots also have the ability to learn from previous interactions and adapt accordingly. This means that they become more efficient over time, reducing the workload for human agents and allowing them to focus on more complex tasks that require a personal touch.

Furthermore, AI technology is constantly evolving, meaning that there are always new ways in which it can be used to improve customer service. From personalized product recommendations based on browsing history to using facial recognition software for seamless checkout experiences, the possibilities are endless. As such, we can expect to see an even greater role for AI in Ecommerce customer service in the future.

Author: Ali Ahmad

Source: Datafloq