An Odyssey through MSc Waters
-
- 0 Rating
- 0 Reviews
- 2 Students Enrolled
An Odyssey through MSc Waters
This courselet offers an in-depth analysis of MSc theses from the LvB Chair of Statistics at HU Berlin. Employing Latent Dirichlet Allocation, we have identified and explored different topics within the theses. We invite you to explore the results of our investigation.
-
- 0 Rating
- 0 Reviews
- 2 Students Enrolled
Requirements
- It is recommended to have a general understanding of: 1. Webscraping 2. Text preprocessing: cleaning data and making corpus 3. Dimenionsionality reduction methods, esp. UMAP 4. Hyperparameter tuning: grid search 5. Topic modeling approaches, esp. LDA
General Overview
Description
This courselet covers:
- Topic Modeling Approaches: An overview of LSA, PLSA, and LDA.
- Webscraping: Collecting MSc theses from HU Berlin website.
- Text Cleaning: Cleaning data from noise, stopwords, and rare words.
- Corpus Creation: Reorganizing data into a suitable format.
- LDA Application: Employing LDA with gridsearch for topic exploration.
- UMAP Visualization: Uncovering visual patterns in text data through UMAP.
- Dynamic Topic Modeling: Exploring topic evolution over time using DTM.
Meet the instructors !
About the Instructor
Hello! I am a member of LDA MSc Theses team in DEDA class, we want to upload our final slides