Text Mining @ TAU
Course page for teaching materials, 24/25 (Semester A)
Text Mining @ TAU
This semester of Text Mining focused on practical NLP workflows, from preprocessing and classical representations through embeddings, transformers, and LLM-based applications.
Course Outline
Text Mining (Semester A) introduces automated analysis of large collections of text, with emphasis on social media and user-generated content. The course covers both unsupervised methods for discovering patterns, trends, and topics, and supervised methods for classifying text into predefined categories. Key applications include sentiment analysis (positive/negative/neutral), basic NLP pipelines, and presenting text-based findings through clear visualizations.
| Lecture | Topic | Focus |
|---|---|---|
| Block 1 | Text mining overview and use cases | What text mining is, where text data comes from (social media, reviews, comments), and how to frame business and research questions. |
| Block 2 | Text data preparation and representation | Cleaning and structuring text data; turning text into analyzable formats; exploratory analysis and simple visual summaries. |
| Block 3 | Unsupervised text analysis: trends and topics | Finding patterns without labels, including trend detection and summarizing main themes in a corpus. |
| Block 4 | Feature engineering for text | Building useful text features (e.g., term weights such as TF-IDF) and understanding their strengths and limitations. |
| Block 5 | Supervised text classification | Training models to assign documents to predefined classes; basic evaluation and practical modeling considerations. |
| Block 6 | Sentiment analysis | Methods and workflows for detecting polarity (positive/negative/neutral) and interpreting results in real applications. |
| Block 7 | Basic NLP models and pipelines | Core NLP building blocks used in modern text analysis and how to combine them into end-to-end solutions. |
| Block 8 | Communicating results with visuals | Presenting text mining outputs clearly: charts and summaries for trends, topics, and classification outcomes. |
Course Materials
Week 0
- Text Mining 24/5 - S0: Course Intro - Week 0 lecture slides
Week 1
- Text Mining 24/5 - S1: Text Processing - Week 1 lecture slides
Week 2
- Text Mining 24/5 - S2: Word2Vec - Week 2 lecture slides
Week 3
- Text Mining 24/5 - S3: Unsupervised Text Mining & Visualization - Week 3 lecture slides
Week 4
- Text Mining 24/5 - S4: Supervised Text Mining - Week 4 lecture slides
Week 5
- Text Mining 24/5 - S5: Attention to SBERT - Week 5 lecture slides
Week 6
- Text Mining 24/5 - S6: LLMs - Week 6 lecture slides