# US 9,519,917 B2 · Context-based advertising > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,519,917 B2 - Document ID: US9519917B2 - Status: Granted - Kind: B2 - Application number: 14/557,173 - Filed: 2014-12-01 - Published: 2015-03-26 - Granted: 2016-12-13 - Earliest claimed: 2007-11-27 - Assignees: eBay Inc. - Applicants: eBay Inc. - Inventors: Thomas Geiger - Web page: https://patentgrep.com/patent/US9519917B2 ## CPC - G06Q 30/0252: {based on events or environment, e.g. weather or festivals} - G06F 40/295: Named entity recognition - G06Q 30/02: Marketing; Price estimation or determination; Fundraising - G06Q 30/0256: {User search} - G06Q 30/0263: {based upon Internet or website rating} - G06F 40/205: Parsing - G06F 40/263: Language identification - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A method and a system for context-based real-time advertising are provided. In example embodiments, a document content that is displayed to a user may be analyzed and keywords may be identified. Selected listings from a publication system may be received; the selected listings may be selected using the keywords. The system may detect user events associated with the keywords and, in response to the detection of the user events, display information related to the listings while maintaining the displaying of the document content. ## Claims ### Claim 1 A method comprising: displaying a document that includes a script and a program, the program, comprising of a programming language and being executable to initiate an analysis that identifies a word in the document as a contextual keyword of the document, the script, comprising of a scripting language and being executable to highlight any contextual keywords identified in the document; by execution, at a client machine, of the program included within the document, initiating the analysis in response to the program sending a request to a server machine; by execution, at the client machine, of the program included within the document, performing the analysis by: determining an occurrence frequency of the word in the document; determining that the word is among a group of most frequently occurring words in the document, the group of most frequently occurring words having a predefined size; identifying an ad category with a highest probability of containing the word based on the word being among the group of most frequently occurring words in the document; comparing the word in the document with words contained in the ad category; and identifying the word in the document as the contextual keyword of the document that is being displayed; selecting an ad from a publication system and from a further ad category that includes the highest number of ads related to the contextual keyword of the document that is being displayed, the selecting of the ad being performed by a processor of the server machine; by execution of the script included within the document, highlighting the word identified as the contextual keyword of the document that is being displayed; detecting, at the client machine, a mouseover event over the word that is highlighted in the document that is being displayed; and displaying, at the client machine, the ad within a tooltip window over the document in response to the detecting of the mouseover event over the word highlighted in the document that is being displayed. ### Claim 2 (depends on 1) The method of claim 1, wherein: the program included within the document is further executable to determine whether any words in the document have been identified as contextual keywords of the document; the method further comprises by execution of the program included within the document, determining that no words in the document have been identified as contextual keywords of the document; and wherein the initiating of the analysis is based on a determination that no words in the document have been identified as contextual keywords of the document. ### Claim 3 (depends on 1) The method of claim 1, wherein: the analysis of the document includes detecting a language of the document. ### Claim 4 (depends on 1) The method of claim 1, wherein: the analysis of the document includes removing one or more words from the document. ### Claim 5 (depends on 1) The method of claim 1, wherein: the identifying of the word in the document as the contextual keyword of the document is in response to a request made by an owner of the document. ### Claim 6 A system comprising: a client machine configured to: display a document that includes a script and a program, the program being executable to initiate an analysis that identifies a word in the document as a contextual keyword of the document, the script being executable to highlight any contextual keywords identified in the document; by executing the program included within the document, initiate the analysis in response to the program sending a request to a server machine; by executing the program included within the document, perform the analysis by: determining an occurrence frequency of the word in the document; determining that the word is among a group of most frequently occurring words in the document, the group having a predefined size; identifying an ad category with a highest probability of containing the word based on the word being among the group of most frequently occurring words in the document; comparing the word in the document with words contained in the ad category; and identifying the word in the document as the contextual keyword of the document that is being displayed; and the server machine comprising a processor and configured to: select an ad from a publication system and from a further ad category that includes the highest number of ads related to the contextual keyword of the document that is being displayed; the client machine including: an event detector configured to detect a mouseover event over the word that is highlighted in the document that is being displayed a user interface configured to display the ad within a tooltip window over the document in response to the mouseover event over the word highlighted in the document that is being displayed. ### Claim 7 (depends on 6) The system of claim 6, wherein: the program included within the document is further executable to determine whether any words in the document have been identified as contextual keywords of the document; the client machine is further configured to, by execution of the program included within the document, determine that no words in the document have been identified as contextual keywords of the document; and the initiating of the analysis is based on a determination that no words in the document have been identified as contextual keywords of the document. ### Claim 8 (depends on 6) The system of claim 6, wherein the client machine is configured to perform at least one of: by executing the program included within the document, detect a language of the document, or by executing the program included within the document, remove one or more words from the document. ### Claim 9 A non-transitory machine-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: displaying a document that includes a script and a program, the program, comprising of a programming language and being executable to initiate an analysis that identifies a word in the document as a contextual keyword of the document, the script, comprising of a scripting language and being executable to highlight any contextual keywords identified in the document; by execution of the program included within the document, initiating the analysis in response to the displaying of the document; by execution of the program included within the document, performing the analysis by: determining an occurrence frequency of the word in the document; determining that the word is among a group of most frequently occurring words in the document, the group of most frequently occurring words having a predefined size; identifying an ad category with a highest probability of containing the word based on the word being among the group of most frequently occurring words in the document; comparing the word in the document with words contained in the ad category; identifying the word in the document as the contextual keyword of the document that is being displayed selecting an ad from a publication system and from a further ad category that includes the highest number of ads related to the contextual keyword of the document that is being displayed; by execution of the script included within the document, highlighting the word identified as the contextual keyword of the document that is being displayed; detecting a mouseover event over the word that is highlighted in the document that is being displayed; and displaying the ad within a tooltip window over the document in response to the detecting of the mouseover event over the word highlighted in the document that is being displayed.