Learning a ranking model using interactions of a user with a jobs list

Patent metadata

Field	Value
Publication number	US-2017221007-A1
Application number	US-201715487015-A
Country	US
Kind code	A1
Filing date	Apr 13, 2017
Priority date	Jun 30, 2015
Publication date	Aug 3, 2017
Grant date	—

Abstract

Learning to rank modeling in the context of an on-line social network is described. A learning to rank model can learn from pairwise preference (e.g., job posting A is more relevant than job posting B for a particular member profile) thus directly optimizing for the rank order of job postings for each member profile. With ranking position taken into consideration during training, top-ranked job postings may be treated by a recommendation system as being of more importance than lower-ranked job postings. In addition, a learning to rank approach may also result in an equal optimization across all member profiles and help minimize bias towards those member profiles that have been paired with a larger number of job postings.

First claim

1 . A computer-implemented method comprising:
collecting training data, the training data comprising a plurality of job lists, each job list from the plurality of job lists comprising respective identifications of a plurality of job postings, each identification of a job posting from the plurality of job postings assigned a relevance label indicating a grade of relevance of that job posting with respect to a member profile associated with that job list, the plurality of job postings maintained by an on-line social network system, the member profile being from a plurality of member profiles representing respective members of the on-line social network system. using at least one processor, learning a ranking model using (1) relevance labels from the training data and (2) rank scores calculated for (member profile, job posting) pairs from the training data; accessing a recommended jobs list, the recommended jobs list generated by a retrieval engine for a member profile representing a member in the on-line social network system, executing the ranking model to determine respective rank scores for items in the recommended jobs list, an item in the recommended jobs list representing a job posting maintained by the on-line social network system; causing the items from the recommended jobs list to be presented on a display device in an order based on the determined respective rank scores.

Assignees

• Linkedin Corp

Inventors

• Tang Lijun
• Huang Eric
• Miao Xu
• Zhou Yitong
• Hardtke David
• Young Joel Daniel

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06Q50/01

  Physics · mapped topic

• G06N99/005

  Physics · mapped topic

• G06Q10/1053 Primary

  Employment or hiring · CPC title

• H04L67/306

  User profiles · CPC title