Automatic Curriculum Recommendation for Employees.

We describe a platform for recommending training assets to employees based on interests, career objectives, organizational hierarchy, and stakeholder or peer feedback. The system integrates content-based and interested-based recommendations across multiple data-streams and interaction modalities to arrive at superior recommendations to those based on just content or interests. The training assets span a wide variety of content formats such as blog articles, podcasts, videos, books, and summaries with the added complexity of multiple content providers. The system incorporates a gamut of interactive information such as likes, ratings, comments, and social activity such as sharing to further personalize recommendations to the user. The system has been deployed in a large organization and is continuously improving from user feedback.