Course Outline

1. Basics of Python and Text Analysis (~ 2 weeks)

First steps with Python
Installing and getting started with the toolset
First steps with text analysis
- Simple explorations of word use
- Sentiment analysis
- Basic Tagging
- Counting Words
- Representations of co-occurrence
- Collocations and n-grams

2. Text Classification (~ 2 weeks)

Classification Methods
- Decision Trees
- Naive Bayes
- Support Vector Machines
- Neural Networks
Testing frameworks
Look at some tools with Graphical User Interfaces

3. Negotiating the problems of real data (~1.5 weeks)

Assembling raw data
- From the web: web pages, twitter, etc.
- From PDFs
Cleaning
Normalizing
- Tokenizing
- Stemming
- Lemmatizing
- Regular Expressions

3. Vector space methods (~1.5 weeks)

About bag-of-words approaches
Converting text to a vector
Various representations of vectors
Use of vector representations to measure text similarity
Distributed vector representations (word embeddings)

4. Unsupervised methods (~1.5 Weeks)

Text Clustering
Topic Modeling

5. Toward New Conceptualizations of Methods in LS (1 Week)

A week of readings

Other Topics

Networks