Recommendation search method for search engine, device and computer readable storage medium

• US 9,934,312 B2
• Filed: 09/26/2014
• Issued: 04/03/2018
• Est. Priority Date: 03/26/2012
• Status: Active Grant

• Current Assignee: Shenzhen Shiji Light Speed Information Technology Company Limited
• Original Assignee: Shenzhen Shiji Light Speed Information Technology Company Limited
• Inventors: He, Jun, Kuang, Jianfeng, Pan, Huajie
• Primary Examiner(s): Ho, Binh V
• Application Number:
  US14/497,993
• Publication Number:
  US 20150012519A1
• Time in Patent Office: 1,285 Days
• Field of Search: 707706
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/90324 using system suggestions
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9532 Query formulation
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A recommendation search method is provided for a search engine. The method includes recording a search log and a click log for query terms during search processes, and determining weights of tendencies of vertical searches for the query terms based on the recorded search log and click log. The method also includes, upon detecting a query term in a search box, determining recommended query terms relating to the query term in the search box, searching for the weights of tendencies of the vertical searches for each of the recommended query terms and determining a tendentious vertical search type for each of the recommended query terms based on the weights of tendencies of the vertical searches, and displaying each of the recommended query terms and links of the tendentious vertical search types for the recommended query terms. Further, the method includes, upon detecting a click on a link of one of the tendentious vertical search types corresponding to one of the recommended query terms, performing a search based on the one of the recommended query terms by using a vertical search engine corresponding to the one of the tendentious vertical search type, and displaying search results.

Claims

1. A recommendation search method for reducing human-machine interactions during a search, comprising:

recording a search log and a click log for query terms during search processes performed with a search engine, the search engine including at least a vertical search engine and a web search engine;

determining, by a processor of the search engine, weights of tendencies of vertical searches for the query terms based on the recorded search log and click log;

upon detecting a query term in a search box, determining, by the processor, recommended query terms relating to the query term in the search box, searching for the weights of tendencies of the vertical searches for each of the recommended query terms and determining a tendentious vertical search type for each of the recommended query terms based on the weights of tendencies of the vertical searches, and displaying each of the recommended query terms and links of the tendentious vertical search types for the recommended query terms;

upon detecting a click on a link of one of the tendentious vertical search types corresponding to one of the recommended query terms, performing, by the processor, a search based on the one of the recommended query terms by using a vertical search engine corresponding to the one of the tendentious vertical search type, and displaying search results,

wherein;

the recording the search log for the query terms further includes;

for each type of vertical search engine, when a search is performed with a vertical search engine of the type of vertical search engine, recording, by the vertical search engine, content of the query term being entered, the number of links being clicked among the search results, and the type of the vertical search engine, and

the recording the click log for the query terms further includes;

recording, by the web search engine, the content of the query term being entered, links being clicked among the search results, and the vertical search types corresponding to the links being clicked, when a search is performed with the web search engine,

the determining weights of tendencies of vertical searches for the query terms based on the recorded search log and click log further includes;

determining weights of the tendencies of the vertical search types for each of the query terms, a weight of tendency of a certain vertical search type being determined by;

determining a first weight of the tendency of the certain vertical search type for a certain query term based on the search log,

determining a second weight of the tendency of the certain vertical search type for the certain query term based on the click log, and

obtaining the weight of the tendency of the certain vertical search type for the certain query term by combining the first weight of the tendency and the second weight of the tendency.

2. The method of claim 1, wherein determining a first weight of the tendency of the certain vertical search type for a certain query term based on the search log further includes:

determining a search weight X/Y, wherein X indicates a number of searches based on the certain query term performed by the certain type of vertical search engine during a specified period of time, and Y indicates a number of all searches performed by the certain type of vertical search engine during the specified period of time;

determining a click weight N/M, wherein N indicates a number of clicked links among the search results for the certain query term of the certain type of vertical search engine during the specified period of time, and M indicates a number of links that have been clicked among all search results in a page of the certain type of vertical search engine during the specified period of time; and

obtaining the first weight of the tendency of the certain type of vertical search for the certain query term by performing a weighted calculation on the search weight and the click weight.

3. The method of claim 2, wherein determining a second weight of the tendency of the certain vertical search type for the certain query term based on the click log further includes:

determining the second weight of the tendency as y/x, wherein x indicates a number of searches based on the certain query term performed by the web search engine during the specified period of time, and y indicates a number of links corresponding to the certain type of vertical search that have been clicked among web search results for the certain query term.

4. The method of claim 3, wherein obtaining the weight of the tendency of the certain vertical search type for the certain query term by combining the first weight of the tendency and the second weight of the tendency further includes:

multiplying the first weight of the tendency with the second weight of the tendency to obtain the weight of the tendency of the certain vertical search type for the certain query term.

5. The method of claim 2, wherein obtaining the first weight of the tendency of the certain type of vertical search for the certain query term by performing a weighted calculation on the search weight and the click weight comprises:

obtaining the first weight of the tendency by utilizing an expression (α

×

X/Y+β

×

N/M), wherein α

indicates a weight coefficient for the search, and β

indicates a weight coefficient for the click.

6. The method of claim 2, wherein obtaining the first weight of the tendency towards the certain type of vertical search for the certain query term by performing a weighted calculation on the search weight and the click weight comprises:

utilizing an expression of (1+X/Y)×

(1+N/M) and normalizing an outcome of the expression to a float number in a range of (0,

     1), wherein the float number is the first weight of the tendency.

7. The method of claim 1, further comprising:

after determining weights of tendencies of vertical searches for the query terms based on the recorded search log and click log, filtering the weights of the tendencies by;

comparing each of the weights of the tendencies of the vertical searches for the query terms with a set threshold; and

when the weight of the tendency of a certain type of vertical search is lower than the threshold, setting the weight of the tendency of the certain type of vertical search to 0,

wherein determining the tendentious vertical search type for each of the recommended query terms based on the weights of tendencies comprises;

when weights of the tendencies of all types of vertical searches for the recommended query term are 0, determining that the recommended query term does not have a tendentious vertical search type;

when weights of the tendencies of all types of vertical searches for the recommended query term are not 0, determining the vertical search type with a highest weight of tendency for the recommended query term as the tendentious vertical search type for the recommended query term.

8. The method of claim 1, wherein determining the tendentious vertical search type for each of the recommended query terms based on the weights of tendencies comprises:

determining the vertical search type with a highest weight of tendency for the recommended query term as the tendentious vertical search type for the recommended query term.

9. The method of claim 1, further comprising:

upon detecting a certain one of the recommended query terms is clicked, searching for contents matching the recommended query term by using a web search engine and displaying the search results.

10. A recommendation search apparatus for reducing human-machine interactions during a search, comprising:

a memory;

a processor coupled to the memory; and

program modules stored in the r memory to be executed by the processor, the program modules comprising;

a logging module for recording a search log and a click log for query terms during search processes performed with a search engine, the search engine including at least a vertical search engine and a web search engine;

a tendency analyzing module for determining weights of tendencies of vertical searches for the query terms based on the recorded search log and the click log;

a query term recommending module for detecting the query terms in a search box and, upon detecting a query term in the search box, determining recommended query terms for the query term;

a vertical search tendency recommending module for searching for the weights of tendencies of the vertical searches for each of the recommended query terms and determining a tendentious vertical search type for each of the recommended query terms based on the weights of tendencies of the vertical searches;

a recommendation display module for displaying each of the recommended query terms and links of the tendentious vertical search types for the recommended query terms; and

a responding module for, upon detecting a click on a link of one of the tendentious vertical search types corresponding to one of the recommended query terms, performing a search based on the one of the recommended query term by using a vertical search engine which corresponds to the one of the tendentious vertical search type, and displaying search results,

wherein;

the recording the search log for the query terms further includes;

for each type of vertical search engine, when a search is performed with a vertical search engine of the type of vertical search engine, recording, by the vertical search engine, content of the query term being entered, the number of links being clicked among the search results, and the type of the vertical search engine, and

the recording the click log for the query terms further includes;

recording, by the web search engine, the content of the query term being entered, links being clicked among the search results, and the vertical search types corresponding to the links being clicked, when a search is performed with the web search engine,

the determining weights of tendencies of vertical searches for the query terms based on the recorded search log and click log further includes;

determining weights of the tendencies of the vertical search types for each of the query terms, a weight of tendency of a certain vertical search type being determined by;

determining a first weight of h tendency of the certain vertical search type for a certain query term based on the search log,

determining a second weight of the tendency of the certain vertical search type for the certain query term based on the click log, and

obtaining the weight of the tendency of the certain vertical search type for the certain query term by combining the first weight of the tendency and the second weight of the tendency.

11. The apparatus of claim 10, wherein determining a first weight of the tendency of the certain vertical search type for a certain query term based on the search log further includes:

determining a search weight X/Y, wherein X indicates a number of searches based on the certain query term performed by the certain type of vertical search engine during a specified period of time, and Y indicates a number of all searches performed by the certain type of vertical search engine during the specified period of time;

determining a click weight NM, wherein N indicates a number of clicked links among the search results for the certain query term of the certain type of vertical search engine during the specified period of time, and M indicates a number of links that have been clicked among all search results in a page of the certain type of vertical search engine during the specified period of time; and

obtaining the first weight of the tendency of the certain type of vertical search for the certain query term by performing a weighted calculation on the search weight and the click weight.

12. The apparatus of claim 11, wherein determining a second weight of the tendency of the certain vertical search type for the certain query term based on the click log further includes:

determining the second weight of the tendency as y/x, wherein x indicates a number of searches based on the certain query term performed by the web search engine during the specified period of time, and y indicates a number of links corresponding to the certain type of vertical search that have been clicked among web search results for the certain query term.

13. The apparatus of claim 12, wherein obtaining the weight of the tendency of the certain vertical search type for the certain query term by combining the first weight of the tendency and the second weight of the tendency further includes:

multiplying the first weight of the tendency with the second weight of the tendency to obtain the weight of the tendency of the certain vertical search type for the certain query term.

14. The apparatus of claim 11, wherein obtaining the first weight of the tendency of the certain type of vertical search for the certain query term by performing a weighted calculation on the search weight and the click weight comprises:

obtaining the first weight of the tendency by utilizing an expression (α

×

X/Y+β

×

N/M), wherein α

indicates a weight coefficient for the search, and β

indicates a weight coefficient for the click.

15. The apparatus of claim 11, wherein obtaining the first weight of the tendency towards the certain type of vertical search for the certain query term by performing a weighted calculation on the search weight and the click weight comprises:

utilizing an expression of (1+X/Y)×

(1+N/M) and normalizing an outcome of the expression to a float number in a range of (0,

     1), wherein the float number is the first weight of the tendency.

16. The apparatus of claim 10, wherein the vertical search tendency recommending module is further configured for:

after determining weights of tendencies of vertical searches for the query terms based on the recorded search log and click log, filtering the weights of the tendencies by;

comparing each of the weights of the tendencies of the vertical searches for the query terms with a set threshold; and

when the weight of the tendency of a certain type of vertical search is lower than the threshold, setting the weight of the tendency of the certain type of vertical search to 0,

wherein determining the tendentious vertical search type for each of the recommended query terms based on the weights of tendencies comprises;

when weights of the tendencies of all types of vertical searches for the recommended query term are 0, determining that the recommended query term does not have a tendentious vertical search type;

when weights of the tendencies of all types of vertical searches for the recommended query term are not 0, determining the vertical search type with a highest weight of tendency for the recommended query term as the tendentious vertical search type for the recommended query term.

17. The apparatus of claim 10, wherein determining the tendentious vertical search type for each of the recommended query terms based on the weights of tendencies comprises:

determining;

the vertical search type with a highest weight of tendency for the recommended query term as the tendentious vertical search type for the recommended query term.