Pattern Recognition: The Powerhouse Behind LLMs and all Real-Time AI

Real-time applications of Artificial Intelligence (AI) and Machine Learning (ML) are transforming industries, from finance to entertainment. One of the great benefits of this age of Generative AI (large language models and the likes), is that it has opened up people’s imagination. People can now see that many things, almost anything, is and are possible with today’s real-time AI capabilities.

In this article, I would like to show the similarities and therefore analogies between generative AI and two of the most popular applications for real-time AI: real-time fraud detection and real-time recommender systems. While these applications seem distinct, they share some fundamental aspects: recognizing patterns, and then acting on these patterns. Some would joke that fraud detection systems only “recommend that the fraudster is put in jail”. Therefore it is useful to reflect on what that fundamental shared core of these, and many other, use cases actually consists of. In this article, we will argue that pattern recognition is that core.

Pattern recognition is how AI and ML are used to identify trends in historical data, understand these patterns, and then utilize these insights to forecast future behavior. This approach is crucial in both fraud detection and recommender systems, enabling them to deliver real-time, insightful and actionable results. Sometimes the action may be to offer a new previously unknown product to a returning customer (in the case of a recommender system). And sometimes the action may be that all the alarms go off and authorities are notified to forcefully lead the bad guys to a safe place (in the case of a fraud detection system). 

Large language models use Pattern Recognition

Large language models (LLMs) leverage advanced pattern recognition to understand, learn, and generate language. Trained on vast amounts of text data, these models analyze patterns in word usage, sentence structure, context, and relationships between concepts. This training enables them to develop a probabilistic understanding of language, identifying how words and phrases typically interact. 

When faced with a question or prompt, an LLM uses this knowledge to predict the most contextually relevant and coherent response by evaluating patterns similar to those it has encountered during training. By iterating on this process across diverse contexts, LLMs excel at producing nuanced, human-like answers that align with the input’s meaning and intent. In effect, the recognised language pattern is used to predict the most qualitative and accurate responses.

Fighting Fraud with Pattern Recognition

Fraud detection aims to identify and prevent fraudulent transactions or activities. This requires analyzing large historical datasets to spot subtle but repetitive patterns that indicate fraudulent behavior. For instance, an e-commerce platform might analyze user behavior, transaction details, network activity - or even a combination of all of the above - to identify suspicious patterns.

Consider a sudden surge in purchases from a new account using multiple credit cards. This pattern deviates from normal user behavior and raises a red flag for potential fraud. Real-time fraud detection systems leverage pattern recognition to detect such patterns and make instantaneous decisions about blocking new incoming suspicious transactions that display similar patterns as the fraudulent ones that were seen before.

Building Robust Fraud Detection Models Requires:

• Data Quality: High-quality data is essential for training accurate fraud detection models. This data should accurately reflect user preferences and behaviors.
• Feature Engineering: Identifying and selecting relevant features that capture fraudulent patterns is crucial. For example, analyzing ratings and their positive/negative rating distributions can help identify suspicious users.
• Robust Algorithms: Fraud detection models need to be robust to adversarial attacks, where fraudsters try to manipulate the system. Graph representations of the interactions between fraudsters and systems, like a Graph Convolutional Network (GCN) for example, offer a promising approach to learning robust user representations for fraud detection.

Recommending the Perfect Choice with Pattern Recognition

Recommender systems aim to predict user preferences and interests, and will suggest items they might enjoy. This would promote more and more profitable interactions with the provided systems - which could be shopping cart systems, media portals, or other systems that would benefit from a more intimate relationship between the provider and the user. These systems learn from past user interactions, such as purchases, ratings, or browsing history, to identify patterns that indicate user interests.

Imagine a user frequently purchases books in the kids reading book genre and leaves positive reviews for authors with a specific writing style. A simple recommender system can recognize this pattern and recommend other science kids reading books by similar authors. Real-time recommender systems utilize predictive pattern recognition to provide up-to-date suggestions based on the latest user interactions. A sophisticated pattern-based recommender system would learn how specific times and days of the week (eg. mornings just before going to kindergarten, or evenings just before bed), specific computers from which the system would be accessed (eg. home vs. work computers), and real-time stock keeping availability would matter in making the best possible decisions and recommendations.

Effective Recommender Systems Depend on:

• Understanding User Behavior: Accurately modeling user preferences and interests from historical data is essential.
• Capturing Contextual Information: Incorporating contextual data, such as time, location, and device, can improve recommendation relevance. For instance, a travel recommender system can use location and weather data to suggest suitable destinations.
• Exploiting Multimodal Data: Utilizing multimodal data, like text reviews and images, provides a richer understanding of user preferences. Deep learning techniques, particularly Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), have proven effective in handling multimodal data.

The Similarities between LLMs, Fraud Detection and Recommender Systems

At the core of both LLMs, real-time fraud detection and recommender systems lies predictive pattern recognition. Both applications rely on analyzing past data, understanding patterns, and leveraging insights to predict future behavior and take action based on that prediction. This shared foundation highlights the versatility and power of this approach.

Consider the similarities:

• Pattern Recognition: Both applications aim to detect and understand patterns in user behavior and data.
• Predictive Modeling: Both utilize historical data to predict future actions, whether fraudulent transactions or preferred items.
• Real-time Analysis: Both operate in real time, analyzing incoming data streams and generating immediate results.

The Hopsworks AI Lakehouse: The Foundation for Real-time Pattern Recognition Systems

Real-time AI pattern recognition applications, like LLMs, fraud detection and recommender systems, thrive on high-quality, readily accessible data. The Hopsworks AI Lakehouse emerges as a powerful solution, enabling organizations to build and deploy these applications efficiently. The Hopsworks AI Lakehouse is the centralized repository for storing, managing, and analyzing data from diverse sources. It integrates the capabilities of a data lake and a machine learning operations (MLOps) platform , providing a unified platform for data-driven AI initiatives.

It seems obvious at this point, but the Hopsworks AI Lakehouse offers significant benefits for Predictive Pattern Recognition applications of all kinds. It offers:

• Feature Engineering and Model Training: the AI Lakehouse facilitates feature engineering and model training by providing tools for data transformation, feature extraction, and model development.
• Centralized Data Management: the AI Lakehouse provides a single source of truth for all data, simplifying data access and management for all teams that are developing predictive models
• Scalability and Performance: the AI Lakehouse is engineered to handle massive data volumes and supports real-time data processing, essential for real-time AI applications.
• Unified governance: the AI Lakehouse will allow for governance on source data, and provide the required explainability and transparency on the end-result, the predictive pattern recognition system.

Wrapping up

Predictive pattern recognition is a transformative force driving real-time AI applications like LLMs, fraud detection and recommender systems. The Hopsworks AI Lakehouse solution empowers organizations to leverage this power effectively, providing a robust foundation for building and deploying real-time AI solutions.

Hopsworks simplifies the process of:

• Data Ingestion and Management: Streamline the process of ingesting data from multiple sources and managing it centrally.
• Feature Engineering: Provide tools for efficient feature extraction and transformation, enabling the creation of powerful predictive models.
• Model Training and Deployment: Facilitate model training and deployment, making it easier to build and operationalize real-time AI applications.

With the combination of powerful AI algorithms and robust infrastructure, businesses can unlock the full potential of predictive pattern recognition, leading to enhanced security, less fraud, improved user experiences, and increased business value.

I hope this was a useful clarification of how different AI use cases share specific characteristics that are all facilitated by the AI Lakehouse.

Let me know if you would like to discuss!

Cheers

Rik

What VAT Fraud Detection and Contact Tracing have in common

In the previous blogpost we already illustrated in some detail that the contact tracing graph that we built, has a lot of similarities with a product recommendation system graph. We focused on a the Person-Visit-Place triangle that we had built in our Contact Tracing Graph data model, and converted the red and yellow bits into a Person-Purchase-Product triangles.

There is of course another part to the contact tracing graph that is also very interesting: the Person-Meets-Person subgraph. We derived that graph from the original contact tracing graph, by assuming that if a Person had visited a Place at the same time as another person, they would have been likely to have had a meeting there. This Person-Meets-Person subgraph was the basis for most of our graph analytics.

Read more »

Securing a sample fraud graph with Neo4j 4.0

This week, we at Neo4j formally released our brightest and shiniest new version of the Neo4j Graphg Database to the world. It's been an amazing journey to this point, and others have reported on this magnificent piece of engineering in more depth. Take a look at Jim's blogpost, or if you are in a hurry, checkout the graphcast below:

Last week, I started playing around with it myself - by digging up my good old faithful beergraph, and illustrating some of the new features in childproofing exercise for beers. Take a look at that post as well for some giggles. Now in this post, I wanted to essentially do the same thing as I did on the beergraph, but using a Fraud dataset. 

Let's see how that would work.

Read more »

Using Neo4j Bloom for fraud detection, discovery and visualization

Over the past couple of weeks, I have been discussing and showing Neo4j's new Bloom graph discovery and visualization product to everyone that would have a moment to spare. It's soooo much fun to show a tool you love, and Bloom is definitely one of those. And I have also recorded some odf these demo-sessions - you can find part 1, part 2 and part 3 of these recordings on this blog. All of these recordings use my (in)famous Belgian Beergraph dataset - and that's all good fun...

But of course, exploring a beergraph is not really a "business-y" use case. So I decided I would record a Bloom demo using a realistic dataset that centers around using Neo4j for Fraud Detection purposes. You will find all of the important concepts of the Beergraph demos here as well:

• navigating the graph using graph patterns
• using nifty selection / deselection techniques to only show what you need in the graph
• creating better visualizations with colours and icons
• editing the graph straight from the Bloom interface
• creating custom search phrases for business uses and giving them near-natural language graph search capabilities.

Here's the recording:

Read more »

Podcast Interview with Sébastien Heymann, Linkurious

As I am coming up on my 5th anniversary working for Neo4j, I am increasingly happy, proud and thankful for the journey that we had - and the many great people that I have met along the way. One of these people is FINALLY appearing on this podcast, and has a history with this blog every since the VERY first article that I wrote in january 2013: in this article, I showed folks how to load the Belgian Beer Graph into Neo4j using a tool that was actually not intended for this use: Gephi. Many beer (related article)-s later, I am now finally talking to Sébastien Heymann, founder and CEO of Linkurio.us, and one of the main people behind Gephi at the time. Here we go:



Read more »

Podcast Interview with Darko Križić, Prodyna

Another stupidly late podcast publication on my behalf. Somewhere early March (yes, I KNOW - dammit!!!) I had a great conversation with one of our prime Neo4j partners in Germany and across Europe these days, called Prodyna. We did a couple of events together, and I found that some of their thinking and case studies really aligned very well with my own. So we got together for a chat. It's a bit annoying because both of us were referring and looking forward to GraphConnect - and I clearly missed that deadline/timeline. But still wanted to share the conversation... Here it is:
 As per usual, here's the transcript of our conversation:

RVB: 00:02.689 Hello, everyone. My name is Rik, Rik van Bruggen from Neo Technology, and here we are again recording another podcast, a little bit closer to home. It's actually a really special podcast for me because it's exactly two years ago since we started it on request or instigation of my dear friend Michael Hunger. And this week we've invited someone from Germany in order to talk a little bit about of the wonderful things that PRODYNA is doing with Neo4J. And that's Darko Krizic from PRODYNA. Hey Darko, how are you? 

Read more »