LinkedIn Recommender System


LinkedIn mainly used for professional networking, including employers posting jobs and job seekers posting their CVs. The networking site allows members to create profiles and make connections with others as a way of establishing professional relationships. This function allows employers to advertise jobs and search for prospective candidates. In turn, job seekers can also search and view the profiles of hiring companies and individuals via existing connections.

Target Users and their goals:

Target Users: employers, job seekers and recruiters who’s goals are

How can you help them to accomplish their goals?

Better job/professional connections/article recommendations.

Reversing engineering:

Linkedin page consists of articles, events, recommended jobs and new connections. New connections which are recommended looks not always relevant to me (approx. 70% are good). I guess these recommendations are based on the similarity of the keywords in the profiles and on my past views. Article recommendations are not as interesting as on social networks, but they are more professionally related. I always find useful job recommendations, I guess it is build based on my previous search and keywords in my profile.

LinkedIn Recommender System:

LinkedIn uses a collaborative filtering for example, each member’s profile on LinkedIn has a “People Who Viewed This Profile Also Viewed” recommendation module. Similarly, job posting page has a job browsemap. All of the browsemaps are powered by a horizontal collaborative filtering platform called Browsemap. Also LinkedIn uses content-based and demographic based filtering. This type of recommender system creates a user profile based on a learning method to determine items that a particular user would like. For example, the site may utilize a keyword system that suggests items with similar keywords in its description to an item the user has previously purchased.


It is hard to recommend LinkedIn how to improve its recommender system as it is very successful professional network and uses advanced methods and algorithms, but I would suggest to improve its article recommendations system, as by now it seems a bit boring. Interesting content can keep users for longer period of time on the website and it also can facilitate accoumplishing users goals.