Skip to main content

New answers tagged

0 votes

score or cost function for AgglomerativeClustering

Another interesting option is to use the Dasgupta score A cost function for similarity-based hierarchical clustering. For this cost function, you need to have an adjacency matrix $N\times N$ (N=number ...
Yonatan Lourie's user avatar
1 vote

Assigning a new document to a cluster based on keywords extracted and tf-idf

You can use K-Nearest Neighbours classification using direct document vectors (if this is a better representation then I do not see an advantage in using tf-idf keywords) and using the cluster label (...
xabash's user avatar
  • 86
1 vote

Calculating weighted cosine similarity between vectors of words

Cosine similarity is not the best option because the data is not a vector. Better options could be Kullback–Leibler divergence or Hellinger distance. Both algorithms quantify the similarity between ...
Brian Spiering's user avatar
0 votes

Grouping similar classes to improve accuracy, whilst maximising the number of classes

To me it sounds like there are two different requirements for the aggregation. The overall task is to aggregate related classes, in a manner that improves accuracy. Breaking this down into two simpler ...
MuhammedYunus's user avatar

Top 50 recent answers are included