TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Latent Semantic Analysis Tutorial

63 pointsby rouliover 12 years ago

5 comments

beagle3over 12 years ago
State of the art implementation, using random projection for reasonably accurate yet hundreds to thousands of times faster: <a href="http://radimrehurek.com/gensim/" rel="nofollow">http://radimrehurek.com/gensim/</a><p>Random Projection is something you should be aware of if you do any kind of large dimensional modeling. It <i>is</i> magic.
评论 #4595609 未加载
textminerover 12 years ago
Truncated SVD has been a wonderful tool for "cleaning up" pairwise cosine similarity data for text document comparisons, graph/network building (for a visual representation of entities represented by documents, embedded in something like Gephi/Sigma.js/D3), and for item-based recommendation systems.<p>The biggest problems I then run into involves choosing a "k" (the dimensions allowed in your truncation). Have had some thoughts about training this unsupervised method (providing labeled data for what "oughta" be the top nearest neighbors for this particular entity, and optimizing toward that) or building an ensemble method on top of many SVD'd truncated vector spaces (though the combination method is unclear to me-- pick kNN from a linear combination of each model's outcomes? Pick the intersection of each method's k nearest neighbors?)<p>To novices looking at this tutorial: NumPy's a wonderful tool for small toy examples, but at a certain scale you will depend heavily on the sparse matrix formats provided for you by SciPy. (That and random projections should curb any memory problems you'll run into for many vector space-based problems, short of operating at a Google/Yahoo scale, or if your target's TBs of logging data).
评论 #4594925 未加载
elchiefover 12 years ago
FYI<p>What are the differences among latent semantic analysis (LSA), latent semantic indexing (LSI), and singular value decomposition (SVD)?<p><a href="http://stats.stackexchange.com/questions/4735/what-are-the-differences-among-latent-semantic-analysis-lsa-latent-semantic-i" rel="nofollow">http://stats.stackexchange.com/questions/4735/what-are-the-d...</a>
kleibaover 12 years ago
Also: "LSA was patented in 1988 (US Patent 4,839,853) by Scott Deerwester, Susan Dumais, George Furnas, Richard Harshman, Thomas Landauer, Karen Lochbaum and Lynn Streeter."<p><a href="https://en.wikipedia.org/wiki/Latent_semantic_analysis" rel="nofollow">https://en.wikipedia.org/wiki/Latent_semantic_analysis</a>
评论 #4595207 未加载
jackhammer2022over 12 years ago
Also here is a nice paper to get started with LSI : <a href="http://www2.denizyuret.com/ref/berry/berry95using.pdf" rel="nofollow">http://www2.denizyuret.com/ref/berry/berry95using.pdf</a>