Discovering Categories and Topics
Single membership models
Mixed membership models
Setting the Number of Clusters
Validating Unsupervised Methods: How Legislators Present Their Work to Constituents
Validating expressed priorities
Classifying Documents into Known Categories
the scores attached to words must closely align with how
the words are used in a particular context.
Dictionaries should be used with substantial caution, or at least coupled with explicit
The consequence of domain specificity and lack of validation is that most analyses based on
dictionaries are built on shaky foundations.
Supervised Learning Methods
two major advantages over dictionary methods.
supervised learning methods are much easier to validate, with
clear statistics that summarize model performance.
Constructing a training set
Applying a supervised learning model
Applying supervised learning methods: Russian military discourse
Reducing Complexity: From Words to Numbers
Alternative Methods for Representing Text
Measuring Latent Features in Texts: Scaling Political Actors
Supervised and UnSupervised Methods for Scaling