Method for ranking documents in a hyperlinked environment using connectivity and selective content analysis
April 9, 2018. Estimated Expiration Date is calculated based on simple USPTO term provisions. It does not account for terminal disclaimers, term adjustments, failure to pay maintenance fees, or other factors which might affect the term of a patent.Patent ReferencesData management system for building a database with multi-dimensional search tree nodes System for linking hypermedia data objects in accordance with associations of source and destination data objects and similarity threshold without using keywords or link-difining terms Hypertext document retrieving apparatus for retrieving hypertext documents relating to each other Document filtering via directed acyclic graphs Automatically generating a topic description for text and searching and sorting text by topic using the same Information presentation in a knowledge base search and retrieval system Method for providing a substitute for a requested inaccessible object by identifying substantially similar objects using weights corresponding to object features Research mode for a knowledge base search and retrieval system Natural language information retrieval system and method Information retrieval from hierarchical compound documents InventorsApplicationNo. 058577 filed on 04/09/1998US Classes:707/5, Query augmenting and refining (e.g., inexact access)707/3, Query processing (i.e., searching)707/102, Generating database or data structure (e.g., via user interface)715/513, Structured document (e.g., HTML, SGML, ODA, CDA)715/514Hierarchical controlExaminersPrimary: Homere, Jean R.Attorney, Agent or FirmInternational ClassG06F 017/21AbstractIn a computerized method, a set of documents is ranked according to their content and their connectivity by using topic distillation. The documents include links that connect the documents to each other, either directly, or indirectly. A graph is constructed in a memory of a computer system. In the graph, nodes represent the documents, and directed edges represent the links. Based on the number of links connecting the various nodes, a subset of documents is selected to form a topic. A second subset of the documents is chosen based on the number of directed edges connecting the nodes. Nodes in the second subset are compared with the topic to determine similarity to the topic, and a relevance weight is correspondingly assigned to each node. Nodes in the second subset having a relevance weight less than a predetermined threshold are pruned from the graph. The documents represented by the remaining nodes in the graph are ranked by connectivity based ranking scheme.Other References
| |
