Identifying listings of multi-site entities based on user behavior signals
US9015142B2 · kind B2 · utility
Assignee
Inventor
Key dates
| Filing date | Jun 10, 2011 |
| Grant date | Apr 21, 2015 |
| Priority date | — |
| Expiry date | Apr 17, 2032 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F16/29
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying listings of multi-site entities based on user behavior signals. In one aspect, a method exploits user behavioral signals that are strong indicators of whether a search result previously presented for a multi-site-entity search query is, in fact, a relevant result, and affords sufficient weight to these behavioral signals to overcome other signals of relevance for the search result. Two such strong user behavior signals include (i) users invoking an information window from a search result and (ii) users submitting a direction request related to a search result. A black list of search results and/or a search result index can be generated based on how the two strong behavior signals associated with the search results compare to threshold values set for the two strong behavior signals.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.