Method and system to discover and recommend interesting documents

摘要

Disclosed are several examples of systems that can read millions of news feeds per day about topics (e.g., your customers, competitors, markets, and partners), and provide a small set of the most relevant items to read to keep current with the overwhelming amount of information currently available. Topics of interest can be chosen by the user of the system for use as seeds. The seeds can be vectorized and compared with the target documents to determine their similarity. The similarities can be sorted from highest to lowest so that the most similar seed and target documents are at the top of the list. This output can be produced in XML format so that an RSS Reader can format the XML. This allows for easy Internet access to these recommendations.

主权项

1. A method for recommending interesting documents to a user using a computer, the method comprising:
selecting a plurality of seed documents for initiating a document search of a plurality of target documents different from the seed documents; obtaining a seed document vector for each of the plurality of seed documents, each seed document having a seed document identifier; obtaining a target document vector for each of the plurality of target documents, each target document having a target document identifier; comparing each target document vector to each seed document vector to obtain a document similarity value for each comparison, the document similarity value representing the similarity of the terms within the seed document and target document; recording in memory a similarity tuple for each document similarity value, where each similarity tuple includes: 1) the document similarity value, 2) the seed document identifier of the seed document used in that comparison, and 3) the target document identifier of the target document used in that comparison, whereby a similarity tuple is generated for every combination of seed document and target document; sorting the plurality of similarity tuples by the document similarity values to generate an ordered list of similarity tuples, whereby the relationship between the seed document, target document, and document similarity value is preserved for each similarity tuple within the ordered list; and generating and displaying a plurality of recommendations of target documents based on the ordered list of similarity tuples, wherein each recommendation specifies a seed document, a target document, and the document similarity value, whereby the relationship between the seed document, target document, and document similarity value is preserved in each recommendation provided to the user.