Unlock instant, AI-driven research and patent intelligence for your innovation.

Search device, search method and clustering device

A search method and clustering technology, applied in the field of search devices, can solve unrealistic problems

Active Publication Date: 2017-10-13
FUJIFILM BUSINESS INNOVATION CORP
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

From the viewpoint of required time, it is unrealistic to calculate the PPR based on the latest attention of the user every time a search query from the user (hereinafter referred to as "user query") is obtained

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Search device, search method and clustering device
  • Search device, search method and clustering device
  • Search device, search method and clustering device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] basic idea

[0027] The reason why the PPR (Personalized Page Rank) algorithm cannot obtain the importance (page rank value) or ranking of nodes in real time is that the calculation of the Markov chain is performed on the entire large-scale network as an object. It takes a lot of time to perform such calculations for all web pages (www: World Wide Web) on the Internet.

[0028] However, in order to obtain node importance or rankings related to user queries (search requests), it would be sufficient to perform Markov chain calculations using only the parts related to user queries instead of the entire original network. For example, in the case of performing a Markov chain calculation to obtain node importance or ranking related to a query about drug development, the range of the Markov chain (representing walking on the network) includes the range of medical science, drug areas of science, biochemistry, etc., but may not be required to include, for example, areas on aero...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a search device, a search method and a clustering device. The search method includes: acquiring learning data; performing machine learning using the learning data for cluster partitioning, and calculating a steady state of each biased Markov chain of each cluster representing the learning result to obtain and store representing each node on the network The belongingness information of the belongingness of each cluster of the learning result; receiving the search condition from the user; extracting a cluster suitable for the search condition based on the node group matching the search condition; A partial network formed by node groups of the same class; and performing a personalized ranking algorithm operation on the cut out partial network to calculate the importance of each node on the partial network, and generate user-specific search results related to search conditions.

Description

technical field [0001] The invention relates to a search device, a search method and a clustering device. Background technique [0002] Page, L. et al. Stanford Digital Library Technologies Project (1998), [Online], Internet [searched 10 January 2011] (http: / / www-db.stanford.edu / ~backrub / pageranksub.ps ) (Non-Patent Document 1) discloses a web page ranking algorithm that appropriately defines a "page ranking" indicating the importance of each web page (node) from a graph structure formed of web pages and hyperlinks. Schematically, the PageRank algorithm defines the PageRank value of each node according to the "probability" assigned to each node in the steady state of the Markov chain. The algorithm is based on a simulation with a procedure (Markov random walk) in which a human representative walks randomly following links on a graph from node to node. The higher the probability that a node exists, the higher the page rank value of the node becomes. [0003] In addition, N...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F17/30G06N20/10
CPCG06N20/00G06N20/10
Inventor 冈本洋
Owner FUJIFILM BUSINESS INNOVATION CORP