| 摘要 |
A computer-implemented method that includes the actions of receiving a request for data indicative of a user segment of a social networking platform, accessing user data of the social networking platform, identifying a portion of the accessed user data, analyzing contents of the portion of the accessed user data, determining happiness scores for the users in the user segment, identifying a first subset of the users as being happy users, and identifying a second subset of the users as being unhappy users, determining a frequency at which the particular user performs one or more actions, comparing the determined frequencies at which the users in the second subset perform the one or more actions to a threshold frequency, identifying a third subset of the users included in the second subset, and generating a happiness index for the user segment that is based on one or more determined happiness scores. |
| 主权项 |
1. A computer-implemented method comprising:
receiving a request for data indicative of a user segment of a social networking platform, with the user segment being associated with one or more user characteristics; in response to the request, accessing user data of the social networking platform, with an item of the accessed user data being associated with one or more user characteristics indicative of a type of user associated with the item of accessed user data; comparing the one or more user characteristics that define the user segment to user characteristics associated with the accessed user data; identifying, based on comparing, a portion of the accessed user data that pertains to users of the social networking platform who are included in the user segment; analyzing contents of the portion of the accessed user data that pertains to the users who are included in the user segment; determining, based on analyzing, happiness scores for the users who are included in the user segment, with a happiness score for a user being indicative a happiness of the user or an unhappiness of the user; based on the determined happiness scores,
identifying a first subset of the users who are included in the user segment as being happy users, with a happy user being a user of the social networking platform who is associated with a happiness score that exceeds a happiness threshold; andidentifying a second subset of the users who are included in the user segment as being unhappy users, with an unhappy user being a user of the social networking platform who is associated with a happiness score that is less than or equal to the happiness threshold; for a particular user included in the second subset, determining a frequency at which the particular user performs one or more actions to cause generation of user data for the particular user; comparing the determined frequencies at which the users in the second subset perform the one or more actions to a threshold frequency; identifying, based on comparing the determined frequencies, a third subset of the users included in the second subset, with a user in the third subset performing the one or more actions less frequently than the threshold frequency; and generating a happiness index for the user segment that is based on one or more determined happiness scores for i) an identified number of users in the first subset, and for ii) a weighted number of users in the second subset, wherein the weighted number of users in the second subset is based at least in part on a number of users in the third subset. |
