Showing 1–2 of 2 results for author: Crepaldi, G

Understanding Twitter Engagement with a Click-Through Rate-based Method

Authors: Andrea Fiandro, Jeanpierre Francois, Isabeau Oliveri, Simone Leonardi, Matteo A. Senese, Giorgio Crepaldi, Alberto Benincasa, Giuseppe Rizzo

Abstract: This paper presents the POLINKS solution to the RecSys Challenge 2020 that ranked 6th in the final leaderboard. We analyze the performance of our solution that utilizes the click-through rate value to address the challenge task, we compare it with a gradient boosting model, and we report the quality indicators utilized for computing the final leaderboard. This paper presents the POLINKS solution to the RecSys Challenge 2020 that ranked 6th in the final leaderboard. We analyze the performance of our solution that utilizes the click-through rate value to address the challenge task, we compare it with a gradient boosting model, and we report the quality indicators utilized for computing the final leaderboard. △ Less

Submitted 30 September, 2020; originally announced October 2020.

Comments: 8 pages, 4 figures, 4 tables, Recsys2020 Challenge

Predict your Click-out: Modeling User-Item Interactions and Session Actions in an Ensemble Learning Fashion

Authors: Andrea Fiandro, Giorgio Crepaldi, Diego Monti, Giuseppe Rizzo, Maurizio Morisio

Abstract: This paper describes the solution of the POLINKS team to the RecSys Challenge 2019 that focuses on the task of predicting the last click-out in a session-based interaction. We propose an ensemble approach comprising a matrix factorization for modeling the interaction user-item, and a session-aware learning model implemented with a recurrent neural network. This method appears to be effective in pr… ▽ More This paper describes the solution of the POLINKS team to the RecSys Challenge 2019 that focuses on the task of predicting the last click-out in a session-based interaction. We propose an ensemble approach comprising a matrix factorization for modeling the interaction user-item, and a session-aware learning model implemented with a recurrent neural network. This method appears to be effective in predicting the last click-out scoring a 0.60277 of Mean Reciprocal Rank on the local test set. △ Less

Submitted 8 February, 2020; originally announced February 2020.

Comments: 5 pages, 5 figures