Weight-coefficient-based hybrid information recommendation

Abstract

Historical behavioral information of a user is retrieved, where the historical behavioral data includes data associated to operations performed by the user on a server. Recommended information sets are determined based on the historical behavioral information. A plurality of weight coefficients are generated for the plurality of recommended information sets. A recommendation list is determined based on the plurality of weight coefficients. It is determined whether the recommendation list satisfies a recommendation condition. If the recommendation list satisfies the recommendation condition, a recommendation based on the recommendation list is transmitted to the user device.

What is claimed is: 
     
         1 . A computer-implemented method, comprising:
 retrieving, by the one or more processors, historical behavioral information of a user, the historical behavioral information comprising data associated to operations performed by the user on a server;   determining, by the one or more processors, a plurality of recommended information sets based on the historical behavioral information;   generating, by the one or more processors, a plurality of weight coefficients, each weight coefficient of the plurality of weight coefficients being generated for each recommended information set of the plurality of recommended information sets;   determining, by the one or more processors, a recommendation list based on the plurality of weight coefficients;   determining, by the one or more processors, whether the recommendation list satisfies a recommendation condition; and   in response to determining that the recommendation list satisfies the recommendation condition, transmitting, by the one or more processors, a recommendation based on the recommendation list to the user.   
     
     
         2 . The computer-implemented method of  claim 1 , further comprising:
 categorizing the historical behavioral information into test information and reference information.   
     
     
         3 . The computer-implemented method of  claim 2 , wherein determining the plurality of recommended information sets comprises:
 determining the plurality of recommended information sets through a plurality of recommendation algorithms different from each other, the test information comprising results of a portion of the historical behavioral information analyzed by the plurality of recommendation algorithms.   
     
     
         4 . The computer-implemented method of  claim 2 , wherein whether the recommendation list satisfies the recommendation condition comprises:
 determining an accuracy of the recommendation list based on the reference information;   determining whether the accuracy is greater than a threshold; and   in response to determining that the accuracy is greater than the threshold, determining that the recommendation list satisfies the recommendation.   
     
     
         5 . The computer-implemented method of  claim 4 , further comprising:
 in response to determining that the accuracy is lower than the threshold, adjusting, by the one or more processors, the weight coefficient of each recommended information set of the plurality of recommended information sets.   
     
     
         6 . The computer-implemented method of  claim 5 , wherein adjusting the weight coefficient comprises:
 determining a value of an adjustment to the weight coefficient for each recommended information set based on an iteration information that is related to the reference information; and   adjusting the weight coefficient of each recommended information set based on the determined value of the adjustment.   
     
     
         7 . The computer-implemented method of  claim 6 , wherein the iteration information comprises at least one of a value of a previous adjustment of the weight coefficient of each recommended information set, an accuracy of a recommendation list that was previously determined, and highest accuracy value of a plurality of accuracy values of previously determined recommendation lists. 
     
     
         8 . The computer-implemented method of  claim 7 , further comprising:
 in response to determining that a number of adjustments reaches an adjustment threshold, determining a most accurate recommendation list from the previously determined recommendation lists based on the plurality of accuracy values; and   generating the recommendation based on the most accurate recommendation list.   
     
     
         9 . The computer-implemented method of  claim 1 , wherein each weight coefficient of the plurality of weight coefficients comprises a sub-weight coefficient of each piece of information comprised in the recommended information set based on a weight coefficient of the respective recommended information set and a recommended weight coefficient of each piece of information relative to the recommended information set. 
     
     
         10 . The computer-implemented method of  claim 9 , further comprising:
 determining, for each piece of information, a sum of sub-weight coefficients of the recommended information sets, and using the sum as a total weight coefficient of the particular piece of information; and   determining the recommendation list based on the total weight coefficient of each piece of information.   
     
     
         11 . A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
 retrieving historical behavioral information of a user, the historical behavioral information comprising data associated to operations performed by the user on a server;   determining a plurality of recommended information sets based on the historical behavioral information;   generating a plurality of weight coefficients, each weight coefficient of the plurality of weight coefficients being generated for each recommended information set of the plurality of recommended information sets;   determining a recommendation list based on the plurality of weight coefficients;   determining whether the recommendation list satisfies a recommendation condition; and   in response to determining that the recommendation list satisfies the recommendation condition, transmitting a recommendation based on the recommendation list to the user.   
     
     
         12 . The non-transitory, computer-readable medium of  claim 11 , the operations further comprising:
 categorizing the historical behavioral information into test information and reference information.   
     
     
         13 . The non-transitory, computer-readable medium of  claim 12 , wherein determining the plurality of recommended information sets comprises:
 determining the plurality of recommended information sets through a plurality of recommendation algorithms different from each other, the test information comprising results of a portion of the historical behavioral information analyzed by the plurality of recommendation algorithms.   
     
     
         14 . The non-transitory, computer-readable medium of  claim 12 , wherein whether the recommendation list satisfies the recommendation condition comprises:
 determining an accuracy of the recommendation list based on the reference information;   determining whether the accuracy is greater than a threshold; and   in response to determining that the accuracy is greater than the threshold, determining that the recommendation list satisfies the recommendation.   
     
     
         15 . The non-transitory, computer-readable medium of  claim 14 , the operations further comprising:
 in response to determining that the accuracy is lower than the threshold, adjusting the weight coefficient of each recommended information set of the plurality of recommended information sets.   
     
     
         16 . A computer-implemented system, comprising:
 one or more computers; and   one or more computer memory devices interoperably coupled with the one or more computers and having tangible, non-transitory, machine-readable media storing one or more instructions that, when executed by the one or more computers, perform operations comprising:   retrieving historical behavioral information of a user, the historical behavioral information comprising data associated to operations performed by the user on a server;   determining a plurality of recommended information sets based on the historical behavioral information;   generating a plurality of weight coefficients, each weight coefficient of the plurality of weight coefficients being generated for each recommended information set of the plurality of recommended information sets;   determining a recommendation list based on the plurality of weight coefficients;   determining whether the recommendation list satisfies a recommendation condition; and   in response to determining that the recommendation list satisfies the recommendation condition, transmitting a recommendation based on the recommendation list to the user.   
     
     
         17 . The computer-implemented system of  claim 16 , the operations further comprising:
 categorizing the historical behavioral information into test information and reference information.   
     
     
         18 . The computer-implemented system of  claim 17 , wherein determining the plurality of recommended information sets comprises:
 determining the plurality of recommended information sets through a plurality of recommendation algorithms different from each other, the test information comprising results of a portion of the historical behavioral information analyzed by the plurality of recommendation algorithms.   
     
     
         19 . The computer-implemented system of  claim 17 , wherein whether the recommendation list satisfies the recommendation condition comprises:
 determining an accuracy of the recommendation list based on the reference information;   determining whether the accuracy is greater than a threshold; and   in response to determining that the accuracy is greater than the threshold, determining that the recommendation list satisfies the recommendation.   
     
     
         20 . The computer-implemented system of  claim 19 , the operations further comprising:
 in response to determining that the accuracy is lower than the threshold, adjusting the weight coefficient of each recommended information set of the plurality of recommended information sets.