Closed nmonath closed 10 years ago
Inputs: List of words that appear in 1+ training documents Word2Vec Pre-Trained Model k - number of clusters
Outputs: Mapping from word to cluster ID
Clustering algorithms to try: Kmeans, DBscan
Use hierarchical clustering
Inputs: List of words that appear in 1+ training documents Word2Vec Pre-Trained Model k - number of clusters
Outputs: Mapping from word to cluster ID