INTRODUCTION

PURPOSE

• 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.

Sources

• Julia Silge's Work

  Source	Details
  Evaluation	Includes details on building an STM model
  Details on Training	includes details on parameter tuning

Packages

• STM

Walkthroughs and Execution

• 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.