Min, max, median and mean distances are all commonly used for this kind of problem.
This variety is often symptomatic of the limitations of the clustering algorithm, however. Here are some examples of alternatives for hierarchical clustering: http://www.cs.berkeley.edu/~jordan/papers/hierarchical-dp.pdf http://www.gatsby.ucl.ac.uk/~ywteh/research/npbayes/nips2004a.pdf http://sra.itc.it/people/sriharsha/hierarchical_dirichlet.pdf http://citeseer.ist.psu.edu/mackay94hierarchical.html Jordan has been involved in some very nice genomic work that infers cladistic relationships from genetic data as well. On 4/14/08 3:27 PM, "Sean Owen (JIRA)" <[EMAIL PROTECTED]> wrote: > > [ > https://issues.apache.org/jira/browse/MAHOUT-19?page=com.atlassian.jira.plugin > .system.issuetabpanels:comment-tabpanel&focusedCommentId=12588828#action_12588 > 828 ] > > Sean Owen commented on MAHOUT-19: > --------------------------------- > > This is probably barely relevant but thought I would mention that the > cluster-based recommenders I've used define distance between clusters as > either the minimum distance between any pair of nodes, one from the first > cluster and one from the other. Alternatively you could define it as the max > distance between any pair. Dunno, maybe that idea could translate to this > situation is the distance metric is proving problematic. > >> Hierarchial clusterer >> --------------------- >> >> Key: MAHOUT-19 >> URL: https://issues.apache.org/jira/browse/MAHOUT-19 >> Project: Mahout >> Issue Type: New Feature >> Components: Clustering >> Reporter: Karl Wettin >> Assignee: Karl Wettin >> Priority: Minor >> Attachments: MAHOUT-19.txt, TestBottomFeed.test.png, >> TestTopFeed.test.png >> >> >> In a hierarchial clusterer the instances are the leaf nodes in a tree where >> branch nodes contains the mean features of and the distance between its >> children. >> For performance reasons I always trained trees from the top->down. I have >> been told that it can cause various effects I never encountered. And I >> believe Huffman solved his problem by training bottom->up? The thing is, I >> don't think it is possible to train the tree top->down using map reduce. I do >> however think it is possible to train it bottom->up. I would very much >> appreciate any thoughts on this. >> Once this tree is trained one can extract clusters in various ways. The mean >> distance between all instances is usually a good maximum distance to allow >> between nodes when navigating the tree in search for a cluster. >> Navigating the tree and gather nodes that are not too far away from each >> other is usually instant if the tree is available in memory or persisted in a >> smart way. In my experience there is not much to win from extracting all >> clusters from start. Also, it usually makes sense to allow for the user to >> modify the cluster boundary variables in real time using a slider or perhaps >> present the named summary of neighbouring clusters, blacklist paths in the >> tree, etc. It is also not to bad to use secondary classification on the >> instances to create worm holes in the tree. I always thought it would be cool >> to visualize it using Touchgraph. >> My focus is on clustering text documents for instant "more like this"-feature >> in search engines and use Tanimoto similarity on the vector spaces to >> calculate the distance. >> See LUCENE-1025 for a single threaded all in memory proof of concept of a >> hierarchial clusterer.
