LinkedIn Improves Search Results For Posts – Search Engine Journal

Share on facebook
Share on twitter
Share on linkedin
Share on whatsapp
Share on pinterest

LinkedIn is overhauling its search architecture for posts, which will lead to faster and more relevant results.
A series of improvements to LinkedIn’s search system for posts will allow it to deliver faster and more relevant results within a simplified user interface.
LinkedIn details the journey of developing new system architecture while explaining the complexity of its old system.
Search results for posts were previously served by two indexes – one for posts in LinkedIn’s main feed and one for articles.
The complex nature made it difficult to build upon, so LinkedIn decided to decouple the two indexes. LinkedIn reveals the entire process in excruciating detail in a new blog post.
Here it is summed up in a diagram:
 
Much of the information LinkedIn shared is geared toward software engineers. In this post, I’ll skip over the technical details and explain what the changes mean for regular users on LinkedIn.
To make search results more relevant, LinkedIn set out to create a system that considers the following aspects:
LinkedIn says its new system must also deliver results from diverse sources.
LinkedIn is leveraging several machine learning techniques to satisfy searcher expectations to meet these goals for relevance and diversity.
Further, LinkedIn crowdsourced human ratings for search results and leveraged the data to ensure its new system meets a certain threshold for quality.
LinkedIn notes that the crowdsourced human annotation data also provides valuable training data to improve the ranking of results.
LinkedIn’s new system, powered by machine learning, improves on the old system in the following ways:
LinkedIn says search results delivered by its new system have increased user satisfaction, reflected by a 20% increase in positive feedback.
Results that are highly relevant to a user’s query, which LinkedIn identifies as “pertinent results,” have led to an aggregate click-through rate improvement of over 10%.
Greater diversity of posts from within the searcher’s social network, their geographic location, and their preferred language has led to a 20% increase in messaging within the searcher’s network.
The time it takes to deliver search results has been reduced by ~62ms for Android, ~34ms for iOS, and ~30ms for web browsers.
LinkedIn shares how it will continue to improve search results for posts. Future updates will involve:
See LinkedIn’s full blog post for all the technical details behind these changes.
Source: LinkedIn
Featured Image: /Shutterstock
Matt G. Southern, Senior News Writer, has been with Search Engine Journal since 2013. With a bachelor’s degree in communications, …
Get our daily newsletter from SEJ’s Founder Loren Baker about the latest news in the industry!
Subscribe to our daily newsletter to get the latest industry news.
Subscribe to our daily newsletter to get the latest industry news.

source

Leave a Replay

Sign up for our Newsletter

Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit