Thank you for your answers. I have no particular need. I just thought that there is the fuzzysearch for syntactical proximity search, there may be some implementation (in Lucene or external, easy to include) for a semantical proximity search (LSI or something similar). Just wanted to test it on some pdf's..
Quoting "J. Delgado" <[EMAIL PROTECTED]>: It all depends for what you need it for. BTW, Latent Semantic Analysis (LSA) is a super set of LSI. LSI concentrates on just how to index and search documents in a reduced dimensional (latent) space whether LSA includes a range of possible analysis that can be done on representations in this space. There are other equivalent techniques (e.g. probabilistic LSI) that can be much more efficient. Perhaps the original requester could give us more information about how he intends to use LSI. For example is this for plain "concept" search or for document classification, clustering, automatic query expansion/suggestion, link/topology analysis or for something else? J.D. 2007/1/29, Mario Alejandro M. <[EMAIL PROTECTED]>: > I also research the use of LSA. > > My interest is simply cluster the information. I found that LSA is a way, > but I'm not convinved is the better (also, is very high in CPU and RAM > consumption). > > -- > Mario Alejandro Montoya > MCP > www.paradondevamos.com > !El mejor sitio de restaurantes y entretenimiento de Colombia! > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] Mit freundlichen Grüßen Kind regards Christoph Pächter --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
