Hi there 👋

Welcome to my blog

Building an LLM Chatbot to Query a Neo4j Patents Database

Getting Started First off, here is the link to the chatbot! It allows the user to query a knowledge graph of one week’s worth of USPTO patent applications from early 2023 by providing a natural language interface. I want to give credit where it’s due for the inspiration for this project. Last week at GraphSummit Minneapolis, I saw Michael Moore and Phani Dathar each give a presentation with demos of a retrieval-augmented generation, or RAG, system that queried a knowledge graph using natural language....

September 30, 2023 Â· 7 min Â· 1426 words Â· Chris

Topic Modeling Arxiv Article Abstracts using NMF ... By ChatGPT

Getting Started So I’ll confess, I’m pretty late to the party here. ChatGPT and other generative AI models have taken the popular conciousness by storm while I’ve been sitting around trying to ignore the hype. I’ve seen previous advancements in generative audio and imagery models and shrugged at the relative lack of societally beneficial applications that surfaced. I’ve even been uncomfortable using the term ‘artificial intelligence’ to refer to any system – you’ll notice it doesn’t appear in any of my content – but that may be starting to change....

February 20, 2023 Â· 6 min Â· 1166 words Â· ChatGPT

Descritizing numerical columns for ML with pandas

Getting Started I am a big fan of Python and its powerful data analysis library, pandas. Today, I’d like to share a trick that’s save me tons of time when trying to one hot encode continuous numerical data in pandas. One hot encoding is a technique used to convert categorical variables into 1s and 0s in order to prepare data for machine learning algorithms. Normally, this is done using the get_dummies() function in pandas....

February 11, 2023 Â· 3 min Â· 474 words Â· Chris

BI Project Critiques

Getting started My first post here is a collection of old BI projects. I wavered on whether or not to post these since they’re no longer representative of my current work, but in the end I felt like it could be useful for early career analysts to see how another data professional has developed over the years. In that spirit, I’ll critique each of these projects with an eye to what I would do differently if I were starting the project again from scratch....

January 30, 2023 Â· 5 min Â· 1035 words Â· Chris