- The purpose of this repo is to start working with introductory techniques for topic modeling and to gather resources and documentation on the topic for R and potentially python.
| Source | Details | Local Name |
|---|---|---|
| Evaluation | Includes details on building an STM model. Human readable by Julia Silge | |
| Details on Training | includes details on parameter tuning. Human readable by Julia Silge | |
| STM Vignette | Official vignette by the STM team. | stmVignette.pdf |
| Topic Models | Introduction to Topic Modeling | TopicModels__BleiLafferty2009.pdf |
- First: Get data from 311
- Julia Silge's
- Data preparation(?)
- Cast to Tidy Text format > DFM
- Build TF_IDF model for comparison model
- Build STM Mode
- Preparation: Build a document term matrix
- Topic selection count: We'll need a way to automate this.
- Data preparation(?)