Semi-supervised LDA (Topics-in-set Knowledge)

[taimoorkhan-nlp]

One sentence summary:
This LDA variant gets manual word to topic mapping for the initial seed words and then applies topic modeling to gather other words around it from the given dataset. Unlike the unsupervised default nature of LDA, the seed words fixes their respective topics to their respective concepts.

Programming language (Python, R, ...):

Task Type:

• Data Collection and Retrieval
• Data Preprocessing
• Data Mining
• Analysis
• Data Visualization