Displaying 2 results from an estimated 2 matches for "miniclust".
2016 Mar 10
2
Introduction and Doubts
...cal
agglomerative clustering.
It is faster and produces good results based on various metrics of cluster
quality.
the algorithm works in following way
The first phase of the algorithm is
divisive, as the original data set(in this case, set of search documents to
cluster) is split recursively into miniclusters through successive
binary splits: the algorithm?s second phase is agglomerative since these
miniclusters
are recombined into the final result. Further, the approach induces during
execution a dynamic
hierarchical grid that will better fit the dataset w.r.t. classical grid
approaches that exploit...
2016 Mar 09
3
Introduction and Doubts
Hello All,I am Nirmal Singhania from NIIT University,India.
I am interested in Clustering of search results Topic.
I have been in field of practical machine learning and information
retrieval from quite some time.
I took various courses/MOOC on Information retrieval and Text Mining and
have been working on real life datasets(KDD99,AWID,Movielens).
Because the problems you face in real life ML/IR