As Pinterest undergoes explosive growth in its user base, the content corpus that users contribute to also expands rapidly. The content and the machine-learning algorithms exist in a feedback cycle where content signals power the algorithms and algorithms influence the content that populates the ecosystem. A major algorithm change can significantly impact how content is exposed. Without proactive monitoring in place, the makeup of the exposed content can drift over time, in a direction that may not be the most beneficial to users.
Grace Huang offers a deep dive into the lifecycle of a piece of Pinterest content, from its birth through its growth and finally to its death, and explores how this process has changed over time. Along the way, Grace explains how a panel of metrics and a content-specific experiment framework were developed to help Pinterest gauge its content ecosystem health in different markets, helping answer questions such as how fast can we activate a piece of content absent of historical signals, do we have enough content to support a new market, and how has the makeup of the content corpus shifted over time? Grace concludes by sharing the story of how a cross-functional effort was bootstrapped to ensure that quality content can thrive in the the ecosystem and make its way to pinners who will find it relevant and engaging.
Grace Huang is the technical lead for data science on Discovery at Pinterest, where discovery products like recommendations and personalization are developed. She is passionate about building data science products around machine learning algorithms to ensure a sustainable ecosystem, and drive better experience for Pinterest users.
Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?
Join the conversation here (requires login)
©2017, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • email@example.com
Apache Hadoop, Hadoop, Apache Spark, Spark, and Apache are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries, and are used with permission. The Apache Software Foundation has no affiliation with and does not endorse, or review the materials provided at this event, which is managed by O'Reilly Media and/or Cloudera.