{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/pixie-a-system-for-recommending-3-billion","title":"Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time","arxiv_id":"1711.07601","date":"2017-11-21","proceeding":null,"authors":["Chantat Eksombatchai","Pranav Jindal","Jerry Zitao Liu","Yuchen Liu","Rahul Sharma","Charles Sugnet","Mark Ulrich","Jure Leskovec"],"abstract":"User experience in modern content discovery applications critically depends\non high-quality personalized recommendations. However, building systems that\nprovide such recommendations presents a major challenge due to a massive pool\nof items, a large number of users, and requirements for recommendations to be\nresponsive to user actions and generated on demand in real-time. Here we\npresent Pixie, a scalable graph-based real-time recommender system that we\ndeveloped and deployed at Pinterest. Given a set of user-specific pins as a\nquery, Pixie selects in real-time from billions of possible pins those that are\nmost related to the query. To generate recommendations, we develop Pixie Random\nWalk algorithm that utilizes the Pinterest object graph of 3 billion nodes and\n17 billion edges. Experiments show that recommendations provided by Pixie lead\nup to 50% higher user engagement when compared to the previous Hadoop-based\nproduction system. Furthermore, we develop a graph pruning strategy at that\nleads to an additional 58% improvement in recommendations. Last, we discuss\nsystem aspects of Pixie, where a single server executes 1,200 recommendation\nrequests per second with 60 millisecond latency. Today, systems backed by Pixie\ncontribute to more than 80% of all user engagement on Pinterest.","url_abs":"http://arxiv.org/abs/1711.07601v1","url_pdf":"http://arxiv.org/pdf/1711.07601v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"pixie-a-system-for-recommending-3-billion","repo_url":"https://github.com/jd557/pixie-rust","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.07601","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}