LinkedIn is currently working on a wide spectrum of improvements to enhance the platform’s ability to collide with potential business partners and minimize bias. One of the important steps has already been made. LinkedIn introduced a reworked 'People You May Know' (PYMK) feature. The updated algorithm was developed with the purpose of balancing out the opportunities of all users and stopping particular categories of accounts from dominating the recommendations section.
How PYMK Works
The PYMK feature is based on machine learning algorithms that select appropriate individual recommendations based on various data sources, including the Economic Graph and unique user interactions. Additionally, the AI-driven system analyzes a list of external sociological factors. While all these look very logical, AI can eventually prioritize particular groups of people based on their higher activity. This leads to significantly lower platform usability for people who are “not chosen” by the system. In simple words, it’s not fair that high-profile users appear in other’s recommendations much more often.
Developers decided to address this problem by balancing the frequency of all users’ appearances in the recommendations. The result of this change is a higher potential number of connection opportunities for all users, regardless of how large or small their profiles are.
Developers note that the update has the potential to increase the number of valuable business connections between users. It means a significantly higher economic potential of the platform for its users and, perhaps, a big change for many entrepreneurs. Do you already notice any improvements in the recommendations you get? Share your experience in the comments and tell others about the update in social networks by linking to this article.