I am using Carrot2 to cluster query results from Solr. Is is possible to force (or at least boost) the occurrence of certain words in the labels, in either Lingo, STC or k-means?
With Lingo, this is already possible with the option "Title word boost", which gives more weight to words appearing in the document title. Can this be extended to other words that I can provide?
I imagine it should be at least possible to append the desired words to the string which is being taken by the "Title word boost" option to let the word boost work, but maybe that is not the right approach.
What would be the way to do it?