GENERATING JOB RECOMMENDATIONS USING CO-VIEWERSHIP SIGNALS
An on-line social network system includes or is in communication with a recommendation system that is configured to use signals indicating that multiple members have viewed the same job posting to determine whether a particular job posting should be recommended to a certain member. The recommendation system monitors job view events with respect to job postings and stores the associated viewership data. The viewership data is used to identify a job posting that has been viewed by a member that also viewed a job previously recommended to a subject member and selectively recommend that co-viewed job posting to the subject member.
1 . A computer-implemented method comprising:
monitoring job view events in an on-line social network system, a job view event from the job view events indicating that a member represented by a profile in the on-line social network system viewed, via a user interface provided by the on-line social network system, a job posting provided in the on-line social network system;
determining, based on the job view events, viewership data for a particular job posting, the viewership data comprising a set of references to profiles in the on-line social network system that represent respective members who viewed the particular job posting via a user interface provided by the on-line social network system;
using at least one processor, examining the viewership data associated with the particular job posting to determine that the particular job posting was viewed by a further member who also viewed a recommended job posting, the recommended job posting previously recommended to a member represented by a subject member profile based on a relevance score generated for the recommended job posting with respect to the subject member profile, the relevance score indicating a likelihood that the member applies for a job represented by the recommended job posting;
generating a co-viewership score for a pair comprising the subject member profile and the particular job posting;
selecting, based on the co-viewership score, the particular job posting to be included in a set of recommendations, the set of recommendations comprising references to respective job postings selected for presentation to the member; and
causing presentation, on a display device, of a reference to the particular job posting.
2 . The method of claim 1 , comprising generating the co-viewership score by determining a number of member profiles that are referenced in viewership data associated the particular job posting and are also referenced in viewership data associated the recommended job posting.
3 . The method of claim 1 , wherein the selecting of the particular job posting to be included in the set of recommendations based on the co-viewership score is in response to detecting a new login session commenced with respect to the member represented by the subject member profile.
4 . The method of claim 1 , comprising generating the co-viewership score for the pair comprising the subject member profile and the particular job posting, the generating of the co-viewership score comprising:
accessing a co-viewership graph, the co-viewership graph is a tripartite graph, a first set of nodes in the co-viewership graph comprising nodes representing member profiles in the on-line social network system, a second set of nodes in the co-viewership graph comprising nodes representing those job postings recommended to at least one member represented by a member profile represented by a node from the first set of nodes, and a third set of nodes in the co-viewership graph, the third set of nodes representing job postings associated with respective viewership data that includes a reference to at least one member profile that is also referenced in viewership data associated with a job posting represented by a node in the second set of nodes; and
applying a graph analysis algorithm to the co-viewership graph to derive the co-viewership score for the pair comprising the subject member profile and the particular job posting.
5 . The method of claim 4 , wherein a weight assigned to an edge associated with a node from the first set of nodes and a node from the second set of nodes is calculated based on relevance score calculated for a pair comprising a member profile and a job posting represented by respective nodes from the first set of nodes and the second set of nodes, the relevance score indicating a likelihood that a member represented by the member profile applies for a job represented by the job posting.
6 . The method of claim 4 , wherein a weight assigned to an edge associated with a node from the second set of nodes and a node from the third set of nodes is calculated based on a browsemap score.
7 . The method of claim 3 , wherein the graph analysis algorithm is a random walk algorithm.
8 . The method of claim 1 , wherein the particular job posting is selected based on a job poster value calculated for the particular job posting, the job poster value reflecting a level of engagement of members of the on-line social network system with the particular job posting.
9 . The method of claim 8 , wherein the causing of the presentation of the reference to the particular job posting comprises:
including the reference to the particular job posting into a set of references to recommended job postings, based on availability of a reserved slot, a reserved slot in the set of references to recommended job postings designated to reference job posting having a job poster value less than or equal to an underperformance threshold value.
10 . The method of claim 1 , wherein the determining of the particular job posting to be recommended to the subject member profile, in addition to being based on the co-viewership score, is also based on a relevance value calculated for a pair comprising the subject member profile and the particular job posting, the relevance value is generated based on a member feature vector representing the subject member profile and a job feature vector representing the particular job posting.
11 . A computer-implemented system comprising:
an events monitor, implemented using at least one processor, to monitor job view events in an on-line social network system, a job view event from the job view events indicating that a member represented by a profile in the on-line social network system viewed, via a user interface provided by the on-line social network system, a job posting provided in the on-line social network system;
a viewership data generator, implemented using at least one processor, to determine, based on the job view events, viewership data for a particular job posting, the viewership data comprising a set of references to profiles in the on-line social network system that represent respective members who viewed the particular job posting via a user interface provided by the on-line social network system;
a co-viewership detector, implemented using at least one processor, to examine the viewership data associated with the particular job posting to determine that the particular job posting was viewed by a further member who also viewed a recommended job posting, the recommended job posting previously recommended to a member represented by a subject member profile based on a relevance score generated for the recommended job posting with respect to the subject member profile, the relevance score indicating a likelihood that the member applies for a job represented by the recommended job posting;
a co-viewership score generator, implemented using at least one processor, to generate a co-viewership score for a pair comprising the subject member profile and the particular job posting;
a recommendation generator, implemented using at least one processor, to select, based on the co-viewership score, the particular job posting to be included in a set of recommendations, the set of recommendations comprising references to respective job postings selected for presentation to the member; and
a presentation module, implemented using at least one processor, to cause presentation, on a display device, of a reference to the particular job posting.
12 . The system of claim 11 , wherein the co-viewership score generator is to generate the co-viewership score by determining a number of member profiles that are referenced in viewership data associated the particular job posting and are also referenced in viewership data associated the recommended job posting.
13 . The system of claim 11 , wherein the recommendation generator is to select the particular job posting to be included in the set of recommendations based on the co-viewership score is in response to detecting a new login session commenced with respect to the member represented by the subject member profile.
14 . The system of claim 11 , wherein the co-viewership score generator is to generate the co-viewership score for the pair comprising the subject member profile and the particular job posting, the generating of the co-viewership score comprising:
accessing a co-viewership graph, the co-viewership graph is a tripartite graph, a first set of nodes in the co-viewership graph comprising nodes representing member profiles in the on-line social network system, a second set of nodes in the co-viewership graph comprising nodes representing those job postings recommended to at least one member represented by a member profile represented by a node from the first set of nodes, and a third set of nodes in the co-viewership graph, the third set of nodes representing job postings associated with respective viewership data that includes a reference to at least one member profile that is also referenced in viewership data associated with a job posting represented by a node in the second set of nodes; and
applying a graph analysis algorithm to the co-viewership graph to derive the co-viewership score for the pair comprising the subject member profile and the particular job posting.
15 . The system of claim 14 , wherein a weight assigned to an edge associated with a node from the first set of nodes and a node from the second set of nodes is calculated based on relevance score calculated for a pair comprising a member profile and a job posting represented by respective nodes from the first set of nodes and the second set of nodes, the relevance score indicating a likelihood that a member represented by the member profile applies for a job represented by the job posting.
16 . The system of claim 14 , wherein a weight assigned to an edge associated with a node from the second set of nodes and a node from the third set of nodes is calculated based on a browsemap score.
17 . The system of claim 13 , wherein the graph analysis algorithm is a random walk algorithm.
18 . The system of claim 11 , wherein the particular job posting is selected based on a job poster value calculated for the particular job posting, the job poster value reflecting a level of engagement of members of the on-line social network system with the particular job posting.
19 . The system of claim 18 , wherein the presentation module is to include the reference to the particular job posting into a set of references to recommended job postings, based on availability of a reserved slot, a reserved slot in the set of references to recommended job postings designated to reference job posting having a job poster value less than or equal to an underperformance threshold value.
20 . A machine-readable non-transitory storage medium having instruction data executable by a machine to cause the machine to perform operations comprising:
monitoring job view events in an on-line social network system, a job view event from the job view events indicating that a member represented by a profile in the on-line social network system viewed, via a user interface provided by the on-line social network system, a job posting provided in the on-line social network system;
determining, based on the job view events, viewership data for a particular job posting, the viewership data comprising a set of references to profiles in the on-line social network system that represent respective members who viewed the particular job posting via a user interface provided by the on-line social network system;
examining the viewership data associated with the particular job posting to determine that the particular job posting was viewed by a further member who also viewed a recommended job posting, the recommended job posting previously recommended to a member represented by a subject member profile based on a relevance score generated for the recommended job posting with respect to the subject member profile, the relevance score indicating a likelihood that the member applies for a job represented by the recommended job posting;
generating a co-viewership score for a pair comprising the subject member profile and the particular job posting;
selecting, based on the co-viewership score, the particular job posting to be included in a set of recommendations, the set of recommendations comprising references to respective job postings selected for presentation to the member; and
causing presentation, on a display device, of a reference to the particular job posting.