Natural Language Processing (NLP) as a potential disaster monitoring tool

SparseGPT is a noteworthy one-shot pruning technique that can halve the size of large language models like GPT-3 without adversely affecting accuracy. In this episode, Jon Krohn provides an overview of this development and explains its commercial and environmental implications.

Additional materials: www.superdatascience.com/650

Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast for sponsorship information.

The Super Data Science Podcast: Machine Learning | AI | Success

650: SparseGPT: Remove 100 Billion Parameters but Retain 100% Accuracy

NLP practitioners: this episode is for you. From the awareness of linguistic elements and annotation to getting the necessary people in the room, Vincent Warmerdam presents to Jon Krohn a recipe for a successful project and the open-source NLP tools to get there.

This episode is brought to you by epic LinkedIn Learning instructor Keith McCormick (https://linkedin.com/learning/instructors/keith-mccormick). Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast for sponsorship information.

In this episode you will learn:
• How Vincent came to work with De Speld [08:57]
• Vincent’s role at Explosion [18:59]
• How users can apply spaCy [21:46]
• Prodigy: Annotate training data more efficiently with scripts [26:28]
• How to manage “skill anxiety” with Calmcode [32:32]
• How Vincent fixed bad labels [42:47]
• The value of understanding linguistics for NLP [54:42]
• How to constrain artificial stupidity [1:02:38]

Additional materials: www.superdatascience.com/659

659: Open-Source Tools for Natural Language Processing

Did you know that Google created Tensorflow and Kubernetes?

Or that Solaris invented NFS? That Amazon created S3?

Or that Uber created Apache Spark Horovod?

Some of the key technologies that companies use today were originally created by businesses trying to solve their own internal challenges.

In this episode of the Tech ONTAP Podcast, NetApp TME Rick Huang and Solutions Architect Ken Hillier join us to discuss Rick's new blog on using NetApp AI with Apache Spark Horovod for deep learning and inference use cases.

Episode 360: NetApp AI and Nvidia with Apache Spark Horovod

In this episode, host Jon Krohn explores with his guest Ajay Jain, Co-Founder of Genmo.ai, how creative general intelligence could take the video industry by storm. They also discuss the models that got Genmo to this point, the applications of NeRF, and how understanding human psychology is so essential to developing models that output high-fidelity video.

This episode is brought to you by the Zerve data science dev environment (https://zerve.ai), by Grafbase (https://grafbase.com), the unified data layer, and by Modelbit (https://modelbit.com), for deploying models in seconds. Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast for sponsorship information.

In this episode you will learn:
• About Genmo.ai and the term “creative general intelligence” [03:47]
• Why Ajay started Genmo.ai [09:26]
• The increased performance of multimodal models [21:12]
• All about Denoising Diffusion Probabilistic Models (DDPMs) [31:03]
• The application of Neural Radiance Fields (NeRF) [55:26]
• Predicting pedestrian behavior at Uber [1:01:50]
• How to save money in the process of training models [1:12:42]

Additional materials: www.superdatascience.com/711

711: Image, Video and 3D-Model Generation from Natural Language, with Dr. Ajay Jain

Parsing structured outputs from Large Language Models (LLMs) is a crucial skill when developing AI-powered applications.

It allows developers to extract precise, predictable information from models, transforming unstructured text into structured data.

In LangChain, there are various methods for parsing structured outputs from LLMs, each with its own strengths and use cases.

In this audio, we’ll explore three primary techniques: the with_structured_output method, the PydanticOutputParser class, and the StructuredOutputParser class.

Parsing LLM Structured Outputs in LangChain: A Comprehensive Guide

Live from the QCon San Francisco Conference, we are talking with Denys Linkov, Head of Machine Learning at Voiceflow. Linkov shares insights on using micro metrics to refine large language models (LLMs), highlighting the importance of granular evaluation, continuous iteration, and rigorous prompt engineering to create reliable and user-focused AI systems.

Read a transcript of this interview: https://bit.ly/49tOvt8

Subscribe to the Software Architects’ Newsletter for your monthly guide to the essential news and experience from industry peers on emerging patterns and technologies:
https://www.infoq.com/software-architects-newsletter 

Upcoming Events:

QCon London (April 7-9, 2025)
Discover new ideas and insights from senior practitioners driving change and innovation in software development.
https://qconlondon.com/ 

InfoQ Dev Summit Boston (June 9-10, 2025)
Actionable insights on today’s critical dev priorities.
devsummit.infoq.com/conference/boston2025

InfoQ Dev Summit Munich (Save the date - October 2025)

QCon San Francisco 2025 (17-21, 2025)
Get practical inspiration and best practices on emerging software trends directly from senior software developers at early adopter companies.
https://qconsf.com/ 

InfoQ Dev Summit New York (Save the date - December 2025)


The InfoQ Podcasts: 
Weekly inspiration to drive innovation and build great teams from senior software leaders. Listen to all our podcasts and read interview transcripts:

- The InfoQ Podcast https://www.infoq.com/podcasts/ 
- Engineering Culture Podcast by InfoQ https://www.infoq.com/podcasts/#engineering_culture 
- Generally AI: https://www.infoq.com/generally-ai-podcast/ 

Follow InfoQ:
- Mastodon: https://techhub.social/@infoq 
- Twitter: twitter.com/InfoQ
- LinkedIn: www.linkedin.com/company/infoq
- Facebook: bit.ly/2jmlyG8
- Instagram: @infoqdotcom
- Youtube: www.youtube.com/infoq

Write for InfoQ:Learn and share the changes and innovations in professional software development.
- Join a community of ex
perts.
- Increase your visibility.
- Grow your career.
https://www.infoq.com/write-for-infoq

InfoQ Podcast

Transcript

Denys Linkov on Micro Metrics for LLM System Evaluation

You’re listening to **Weekly AI News with Shaily!** 📈 Hello, AI aficionados, tech trendsetters, and curious minds alike. In our **weekly AI news roundup** for the **week starting October 23, 2023**, we dive into a quintet of AI leaps that are reshaping our world. 🗞️

First up, research labs are innovating like a beehive with new neural network architectures that excel in **image recognition** and **self-driving** technologies! 🚗💨 These breakthroughs promise a revolution in **autonomous vehicles** and healthcare, ensuring precision and trust as we edge toward roads without human drivers. 

Next, we unpack **deep learning breakthroughs** making waves in finance and healthcare, supercharging data analysis for faster insights! 💡🔍 Plus, the U.S. has introduced Executive Order 14179, a policy shake-up to foster innovation and ensure America stays a leader in AI governance. 🏛️

We also discuss the rise of **agentic AI**—systems that strategically operate on their own—and explore enhanced **natural language processing** models redefining human-machine communication. 🗣️

So, what do these advancements mean for researchers, developers, and business leaders? Stay tuned for insights and expert analysis! 🔍✨ 

If you enjoyed this dive into **the latest machine learning developments**, hit that subscribe button, follow me on YouTube, Twitter, LinkedIn, or Medium, and join us next time! Stay curious, stay informed, and keep making money out of data! 💡💰

🚀 Unlock the Future: AI News with Shaily for Week starting 2025-04-21—5 Game-Changing Innovations!

Hey there! 👋 Welcome to *AI with Shaily*, hosted by me, Shailendra Kumar—a friendly guide to the latest and greatest in AI technology. Today, we're exploring some exciting buzz around AI chatbots that are transforming how we search, learn, and connect online. 🤖✨

Not too long ago, chatting with bots was like playing 20 Questions with a parrot—fun but pretty limited. Now, these digital helpers have leveled up big time! They’re no longer just answering simple queries; they’re becoming true conversational partners. They can chat about your hobbies, help troubleshoot tech problems, and even pull in fresh info from the internet in real-time. Imagine ChatGPT, which recently got a major upgrade: it can now combine its built-in knowledge (updated through June 2024) with live web searches. 🌐📅 This means it’s like having a wise friend who reads the morning news before chatting with you, rather than relying on old info.

But here’s a twist to think about: just like your favorite news site or social media feed, these live search features can bring in geographical and cultural biases. 🗺️🌍 So, if you ask for expert opinions, the AI might lean towards sources from certain regions—kind of like how we all have our go-to news outlets back home. It’s natural, but it raises an important question: how do we make sure AI gives us a truly global perspective without creating echo chambers? The good news is that developers are actively working on this, focusing on transparency and giving users more control. 🔍⚖️

On the conversational front, modern AI chatbots are way beyond scripted Q&A or menu-based options. They can hold meaningful conversations about 70% of the time, asking follow-up questions and guiding you until your issue is solved. 🗣️💡 This means less frustration and fewer “Let me connect you to a human agent” moments. It’s like having a customer support rep who actually listens and sticks with you—except it’s AI!

To break it down, chatbot designs generally fall into two camps: the old-school “choose your own adventure” style with menus, and the new wave of intelligent bots powered by natural language processing that understand your questions just as you’d naturally ask them. 🕹️➡️🗨️ This shift makes chatting with bots feel more like talking with a friend—smoother and way more enjoyable.

Why does all this matter? Because knowing these features and quirks helps you become a smarter AI user. When you chat with bots, keep in mind the knowledge cutoff date and the current limits of live search. Be aware of potential cultural biases, and don’t hesitate to ask questions in a conversational way—it works better! 🧠💬 Bonus tip: try rephrasing your questions or specifying the perspective or region you want. That helps the chatbot give you more tailored, relevant answers.

I remember my first experience with ChatGPT’s integrated search. I asked about trending news and got mostly local headlines, which was surprising. But when I adjusted my query to “global tech news,” I got a much richer, broader view. 🌎💻 That moment showed me how a simple tweak can unlock AI’s full potential.

To close, here’s a fitting quote from Alan Turing: *“We can only see a short distance ahead, but we can see plenty there that needs to be done.”* AI chatbots are evolving fast, reshaping how we communicate and discover information. As they grow, so does our need to understand their design and limitations. 🚀🤔

You can find more of my insights on YouTube, Twitter, LinkedIn, and Medium—just search for Shailendra Kumar or *AI with Shaily*. If you enjoyed this segment, please subscribe and share your thoughts—I love hearing from you! 🎥🐦💼✍️

Thanks for tuning into *AI with Shaily*. Until next time, keep asking, keep exploring, and keep the conversation going! 🙌💡

AI Chatbots Unleashed: How Smarter Bots Are Changing the Way We Search and Connect

Welcome to *AI with Shaily*! 🎙️ Hosted by Shailendra Kumar, this show dives deep into the fascinating world of artificial intelligence, bringing listeners the freshest and most exciting developments in the field. Today’s episode focuses on the revolutionary advancements in data science that are reshaping fraud detection, trends expected to lead the industry all the way through 2026. 🚀

Shailendra shares a compelling real-world story from fintech, where a cutting-edge fraud detection system stopped a scammer before any money was lost. The secret? Real-time, self-learning AI that adapts instantly by recognizing unusual patterns as they happen. Unlike traditional, static rule-based systems, this dynamic approach reduces false alarms and manual reviews, creating a faster and more effective shield against cybercrime. 🛡️💡 This innovation has banking and e-commerce sectors buzzing across social media platforms.

The discussion then moves to the power of deep learning and neural networks—described as the “brainy detectives” of AI. These sophisticated transformer models analyze vast, complex datasets to detect subtle fraudulent activities that were previously invisible. They can flag suspicious behavior early, stopping fraudsters in their tracks. Tech communities on LinkedIn and Reddit are abuzz with success stories showcasing how these AI tools are leveling the playing field across various industries. 🧠🔍

An intriguing twist highlights that fraud isn’t limited to just financial transactions—it also hides in language. Enter natural language processing (NLP), which enables AI to scan emails, chats, and even voice calls to spot social engineering and phishing attempts. This multi-channel fraud detection approach is gaining attention on Twitter and call center forums for its ability to catch fraud attempts in real time before they escalate. 🗣️📧🔐

Data integration plays a crucial role too. Combining diverse data sources—like transaction logs, user behavior, and external databases—builds a robust fraud detection fortress stronger than any single source alone. This integrated approach, backed by strong data governance, is a hot topic among data strategists on professional networks. After all, even the smartest AI can falter without clean, unified data. 📊🔗

To process this massive influx of data swiftly, scalable big data processing powered by GPUs is essential. Shailendra likens it to upgrading from a bicycle to a sports car for AI computations, enabling instant fraud detection even when analyzing millions of transactions per second. Cloud communities and AI enthusiasts are actively exchanging tips on optimizing this powerful technology. ⚡🚗💨

For AI practitioners and enthusiasts, Shailendra offers a bonus tip: prioritize flexibility and multi-channel analysis when developing or selecting fraud detection systems. Since fraudsters constantly evolve, defenses must evolve too—monitoring not just transactions, but conversations and behaviors as well. 🔄🎯

The episode closes with a thought-provoking question about balancing rapid fraud detection with fairness—how do we catch fraudsters quickly without wrongly accusing innocent users? Vigilance is vital, but so is justice. ⚖️🤔

Ending on an inspiring note, Shailendra quotes Alan Turing: “We can only see a short distance ahead, but we can see plenty there that needs to be done.” Fraud detection exemplifies this journey—remarkable progress has been made, yet the battle against fraudsters continues. 🖥️✨

Stay connected with Shailendra Kumar on YouTube, Twitter, LinkedIn, and Medium for more insightful AI discussions. Don’t forget to subscribe and share your thoughts—your engagement enriches the conversation. Until next time, keep questioning and keep innovating! 🔍🤖💬

AI Fraud Fighters: How Real-Time Learning and Big Data Are Beating Scammers Before They Strike

Welcome to "AI with Shaily" 🎙️, your ultimate destination for the freshest and most exciting updates in the world of artificial intelligence, hosted by Shailendra Kumar, also known as Shaily. 

In this latest episode, Shaily dives into a fascinating breakthrough from Technion’s Faculty of Data and Decision Sciences. Imagine a technology that can monitor your eye movements 👀 as you read and decode exactly how you’re interacting with the text. Whether you’re deeply understanding the material, skimming for quick facts, or revisiting familiar content, this AI-powered system knows! This incredible innovation is the brainchild of a talented team led by doctoral student Omer Shubi and master’s student Cfir Hadar, guided by Dr. Yevgeni Berzak. By combining machine learning 🤖 and natural language processing 📚, their system analyzes eye movement patterns to understand not just what you read, but how and why you read it.

This groundbreaking research was showcased at the prestigious ACL 2025 conference in Vienna 🇦🇹 and holds immense promise for education and e-learning. Imagine adaptive learning platforms that respond in real-time to your reading behavior, adjusting lessons based on whether you’re struggling or breezing through content. This means personalized education tailored to your unique learning style—no more generic one-size-fits-all approaches!

Shaily shares a personal reflection, recalling tutoring sessions where he noticed students’ glazed eyes without them saying a word. With this new technology, educators could instantly gain those insights, making teaching more dynamic, responsive, and effective 🎓.

For lifelong learners, Shaily offers a handy tip 💡: to improve retention, alternate between deep reading and scanning for key points. This keeps your brain engaged, and as the new research suggests, your eyes reveal a lot about your reading habits.

To leave you with a thought-provoking question: If AI can decode your gaze to understand your thoughts, how comfortable would you feel about having your eye movements tracked by machines? 🤔

Shaily wraps up with a nod to the legendary Alan Turing’s famous question, “Can machines think?” With innovations like this eye-tracking AI, the answer seems to be edging closer to yes.

Don’t miss out on more AI insights—follow Shaily on YouTube, Twitter, LinkedIn, and Medium. Subscribe to "AI with Shaily" for your weekly dose of AI news and share your thoughts in the comments: Are you excited or a bit spooked by this eye-tracking technology? 

This is Shailendra Kumar signing off—keep those eyes curious and your mind open! 👓✨

Eye-Tracking AI: How Machines Decode Your Reading Habits in Real Time

Unlock the secrets of data mastery with Shailendra, a 27-year veteran in AI, ML, and Data Science, driving AU$ 6 Billion in transformative value. As the author of 'Making Money out of Data,' their strategies redefine digital success. With expertise honed at IBM and Accenture, they drive innovation and data monetization. Explore how loyalty programs like Flybuys and Everyday Rewards revolutionize business—connect with Shaily today.

Each week, the Tech ONTAP Podcast dives into all-things NetApp, including storage, public & private cloud, and much more. The team will also be interviewing subject-matter experts from across the industry and detailing various best practices for getting the most out of your datacenter.

Software engineers, architects and team leads have found inspiration to drive change and innovation in their team by listening to the weekly InfoQ Podcast. They have received essential information that helped them validate their software development map. We have achieved that by interviewing some of the top CTOs, engineers and technology directors from companies like Uber, Netflix and more. Over 1,200,000 downloads in the last 3 years.

The latest machine learning, A.I., and data career topics from across both academia and industry are brought to you by host Dr. Jon Krohn on the Super Data Science Podcast.

As the quantity of data on our planet doubles every couple of years and with this trend set to continue for decades to come, there's an unprecedented opportunity for you to make a meaningful impact in your lifetime. In conversation with the biggest names in the data science industry, Jon cuts through hype to fuel that professional impact.

Whether you're curious about getting started in a data career or you're a deep technical expert, whether you'd like to understand what A.I. is or you'd like to integrate more data-driven processes into your business, we have inspiring guests and lighthearted conversation for you to enjoy.

We cover tools, techniques, and implementation tricks across data collection, databases, analytics, predictive modeling, visualization, software engineering, real-world applications, commercialization, and entrepreneurship − everything you need to crush it with data science.

New tracks tagged #natural language processing