LinkedIn Salary: A System for Secure Collection and Presentation of Structured Compensation Insights to Job Seekers

摘要

Online professional social networks such as LinkedIn have enhanced theability of job seekers to discover and assess career opportunities, and theability of job providers to discover and assess potential candidates. For mostjob seekers, salary (or broadly compensation) is a crucial consideration inchoosing a new job. At the same time, job seekers face challenges in learningthe compensation associated with different jobs, given the sensitive nature ofcompensation data and the dearth of reliable sources containing compensationdata. Towards the goal of helping the world's professionals optimize theirearning potential through salary transparency, we present LinkedIn Salary, asystem for collecting compensation information from LinkedIn members andproviding compensation insights to job seekers. We present the overall designand architecture, and describe the key components needed for the securecollection, de-identification, and processing of compensation data, focusing onthe unique challenges associated with privacy and security. We perform anexperimental study with more than one year of compensation submission historydata collected from over 1.5 million LinkedIn members, thereby demonstratingthe tradeoffs between privacy and modeling needs. We also highlight the lessonslearned from the production deployment of this system at LinkedIn.