The first meta search tool based on Singular Value Decomposition or SVD. SVD is a powerful method for finding patterns in data used in, for example, Latent Semantic Indexing or LSI. iMetaSearch uses SVD for clustering results and building a relevance network that shows how related index words and results are to each other. By clicking an index word or result, you can instantly see other related index words and results.
iMetaSearch gathers results (web pages, blogs, messages, papers, etc) from search engines, does SVD based analysis, and displays them in a fast, no-frills, interactive interface so you can explore the results.
iMetaSearch includes the following features:
1) No limit searches - there are no arbitrary limits on the number of results or on the number of index words,
2) SVD based analysis - automatically generates an index, clusters, and a relevance network among index words and search results,
3) Filter for names and emails ? filter the index to display only those words or phrases that contain likely names or email addresses,
4) Adjustable cluster and index sizes ? the number of clusters and the size of the index are adjustable with slider controls,
5) Export to HTML or CSV - export index words to CSV (for Microsoft Excel) files, or export search results to HTML or CSV files,
6) Follow-up searches - right click menu allows follow-up searches from index words or phrases in one click,
7) Import or build engines - you can import engines easily, or build engines for your favorite sources.
iMetaSearch is designed for quick interactive searching and interactive analysis of search results. Unlike other search tools, the relevance of all index words and search results is updated instantly when you mark an index word or result of interest. You can find related index words or results without having to scan through the entire list of results or words.