Real-time Recommendation of Streamed Data

Abstract

This tutorial addressed two trending topics in the field of recommender systems research, namely A/B testing and realtime recommendations of streamed data. Focusing on the news domain, participants learned how to benchmark the performance of stream-based recommendation algorithms in a live recommender system and in a simulated environment.

Continuous Evaluation of Large-Scale Information Access Systems: A Case for Living Labs

“…We understand that these requirements were new to most researchers and that these different entry requirements might hold them back from participating. We addressed this by offering tutorials (e.g., at ECIR'15 (Hopfgartner and Brodt, 2015) and ACM RecSys'15 (Hopfgartner et al, 2015b)), and by providing detailed instructions on how to get started on the NewsREEL website. • In the online task, participants had to deal with fulfilling two goals at once.…”

Lc‐Stream: An elastic scheduling strategy with latency constraints in geo‐distributed stream computing environments

“…Common big data real-time stream computing systems such as Storm, 4 Spark Streaming 5 and Flink 6 play an important role in many fields, including online system monitoring, 7 mobile sensing data processing 8 and the Internet of Things, 9 financial risk control, 10 recommendation systems, 11 and threat detection. 12 Storm is a relatively low latency real-time distributed stream processing framework that comes with a default scheduling scheme.…”