RecSys 2020 Challenge Workshop: Engagement Prediction on Twitter’s Home Timeline

The workshop features presentations of accepted contributions to the RecSys Challenge 2020, organized by Politecnico di Bari, Free University of Bozen-Bolzano, TU Wien, University of Colorado, Boulder, and Universidade Federal de Campina Grande, and sponsored by Twitter. The challenge focuses on a real-world task of Tweet engagement prediction in a dynamic environment. The goal is to predict the probability for different types of engagement (Like, Reply, Retweet, and Retweet with comment) of a target user for a set of Tweets, based on heterogeneous input data. To this end, Twitter has released a large public dataset of ~160M public Tweets, obtained by subsampling within ~2 weeks, that contains engagement features, user features, and Tweet features. A peculiarity of this challenge is related to the recent regulations on data protection and privacy. The challenge data set was compliant: if a user deleted a Tweet, or their data from Twitter, the dataset was promptly updated. Moreover, each change in the dataset implied new evaluations of all submissions and the update of the leaderboard metrics. The challenge was well received with 1,131 registered users. In the final phase, 20 teams were competing for the winning position. These teams had an average size of approximately 4 participants and developed an overall number of 127 different methods.

[1]  Paul Lamere,et al.  Recsys challenge 2018: automatic music playlist continuation , 2018, RecSys.

[2]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[3]  Quoc V. Le,et al.  Proximal regularization for online and batch learning , 2009, ICML '09.

[4]  Thomas Wolf,et al.  DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter , 2019, ArXiv.

[5]  Tommaso Di Noia,et al.  2nd Workshop on Knowledge-aware and Conversational Recommender Systems - KaRS , 2019, CIKM.

[6]  Peter Knees,et al.  RecSys challenge 2019: session-based hotel recommendations , 2019, RecSys.

[7]  Anna Veronika Dorogush,et al.  CatBoost: unbiased boosting with categorical features , 2017, NeurIPS.

[8]  Tie-Yan Liu,et al.  LightGBM: A Highly Efficient Gradient Boosting Decision Tree , 2017, NIPS.

[9]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[10]  Judith Masthoff,et al.  How to Use Social Relationships in Group Recommenders: Empirical Evidence , 2018, UMAP.

[11]  Xavier Bresson,et al.  Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks , 2017, NIPS.

[12]  Fabian Abel,et al.  RecSys Challenge 2017: Offline and Online Evaluation , 2017, RecSys.

[13]  Alejandro Bellogín,et al.  Time-Aware Novelty Metrics for Recommender Systems , 2018, ECIR.

[14]  Domonkos Tikk,et al.  Alternating least squares for personalized ranking , 2012, RecSys.

[15]  Jessie J. Smith,et al.  Privacy-Preserving Recommender Systems Challenge on Twitter's Home Timeline , 2020, ArXiv.

[16]  Tommaso Di Noia,et al.  Local Popularity and Time in top-N Recommendation , 2018, ECIR.

[17]  Alejandro Bellogín,et al.  Discovering Related Users in Location-based Social Networks , 2020, UMAP.

[18]  Andrea Calì,et al.  Exposing Open Street Map in the Linked Data cloud , 2016, SEBD.