How LinkedIn Uses Machine Learning to Analyze Member Data
Margaretta Colangelo
Leading AI Analyst | Speaker | Writer | AI Newsletter 57,000+ subscribers
A team of 16 scientists and software engineers at LinkedIn are using machine learning techniques to build the LinkedIn knowledge graph (LKG) a universal knowledge base. The knowledge graph is based on user data including jobs, titles, skills, companies, geographic locations, schools etc.
About 450 million people have LinkedIn profiles. Every time they connect with each other, update their profiles, join groups, post jobs, and generate content they create data that is mined and analyzed.
450 million members
190 million job listings
9 million companies
35 thousand skills
28 thousand schools
24 thousand titles
Content Processing Framework
LinkedIn has developed a content processing framework that uses machine learning to infer relationships from this user data. For example, when members enter explicit skills to their profiles a machine learning tool infers other skills.
领英推è
Inferred Relationships
Trillions of member-generated and LinkedIn-inferred relationships exist in the LKG. Analysis of these relationships reveals hidden relationships between members and the strength of these relationships. LinkedIn can also estimate the value that these connections provide to members. The estimated connection strength and value are currently used in various LinkedIn products.
LinkedIn members currently have access to very basic analytics detailing who views their profile and who reads their posts.
Unlocking Value For LinkedIn Members
With these machine learning models, LinkedIn can estimate the value that connections provide to members. Member generated data analyzed with LinkedIn's advanced analytics technology has the potential to unlock tremendous value for LinkedIn members should LinkedIn reveal this data to members.
Comments
I'm interested in your feedback - please leave your comments. To subscribe please click subscribe at the top of this article.
Copyright ? 2016 Margaretta Colangelo. All Rights Reserved.
This article was written by?Margaretta Colangelo.?Margaretta is a leading AI analyst based in San Francisco. She serves on the advisory board of the AI Precision Health Institute at the University of Hawai?i?Cancer Center.?
Twitter?@realmargaretta
An Exponential Leader (Inventor-Innovator-Entrepreneur-Cardiologist) with Moonshots and a Massive Transformative Purpose (MTP - Democratizing Molecular Imaging Globally) to help people live longer and healthier lives.
8 å¹´Interesting!