Building a Semantic Book Recommender with Python, OpenAI, LangChain, and Gradio

Learn to create an advanced book recommendation system in this comprehensive Python programming course that combines large language models, semantic analysis, and modern development tools. Master essential techniques including data preparation, vector search implementation, zero-shot text classification, and sentiment analysis using OpenAI and LangChain. Build a complete recommendation engine from scratch, working with real book datasets, implementing vector databases, and creating an interactive user interface with Gradio. Gain hands-on experience with PyCharm Professional, explore various LLM models on Hugging Face, and learn to evaluate classifier accuracy. Perfect for developers interested in applying natural language processing and machine learning to create practical applications, with access to extensive resources including GitHub code repositories, Kaggle datasets, and supplementary learning materials on transformer models and LLM concepts.

Syllabus

Intro
Introduction to getting and preparing text data
Starting a new PyCharm project
Patterns of missing data
Checking the number of categories
Remove short descriptions
Final cleaning steps
Introduction to LLMs and vector search
LangChain
Splitting the books using CharacterTextSplitter
Building the vector database
Getting book recommendations using vector search
Introduction to zero-shot text classification using LLMs
Finding LLMs for zero-shot classification on Hugging Face
Classifying book descriptions
Checking classifier accuracy
Introduction to using LLMs for sentiment analysis
Finding fine-tuned LLMs for sentiment analysis
Extracting emotions from book descriptions
Introduction to Gradio
Building a Gradio dashboard to recommend books
Outro