TY - GEN
T1 - A-HA
T2 - 2019 ACM Recommender Systems Challenge Workshop, RecSys Challenge 2019, held at the 13th ACM Conference on Recommender Systems, ACM RecSys 2019
AU - Kung-Hsiang, Huang
AU - Tzong-Hann, Lee
AU - Yi-Fu, Fu
AU - Yao-Chun, Chan
AU - Shou-De, Lin
AU - Yi-Ting, Lee
AU - Yi-Hui, Lee
N1 - Publisher Copyright:
© 2019 Association for Computing Machinery.
PY - 2019/9/20
Y1 - 2019/9/20
N2 - Session-based recommender system refers to a specific type of recommender system that focuses more on the transactional structure of each session rather than the user and item interactions [16]. It is stated that the users' interactions are mostly homogeneous in the same sessions, while being heterogeneous across different sessions [5]. Therefore, it is essential to extract the interest dynamics of users within each session. The 2019 ACM Recsys Challenge [10] aims to apply session-based recommender systems to the domain of travel metasearch. The goal is to predict which hotels are clicked in the search results based on the context of each session. In this paper, we propose our approach to effectively tackle the challenge. It involves an ensemble of three models, LightGBM, XGBoost, and a Neural Network based on DeepFM [6] that is capable of handling sequential features. Our team, RosettaAI, won the 4th place in this challenge, scoring 0.679933 on the final leaderboard. The source code is available online 1
AB - Session-based recommender system refers to a specific type of recommender system that focuses more on the transactional structure of each session rather than the user and item interactions [16]. It is stated that the users' interactions are mostly homogeneous in the same sessions, while being heterogeneous across different sessions [5]. Therefore, it is essential to extract the interest dynamics of users within each session. The 2019 ACM Recsys Challenge [10] aims to apply session-based recommender systems to the domain of travel metasearch. The goal is to predict which hotels are clicked in the search results based on the context of each session. In this paper, we propose our approach to effectively tackle the challenge. It involves an ensemble of three models, LightGBM, XGBoost, and a Neural Network based on DeepFM [6] that is capable of handling sequential features. Our team, RosettaAI, won the 4th place in this challenge, scoring 0.679933 on the final leaderboard. The source code is available online 1
UR - http://www.scopus.com/inward/record.url?scp=85076710800&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076710800&partnerID=8YFLogxK
U2 - 10.1145/3359555.3359560
DO - 10.1145/3359555.3359560
M3 - Conference contribution
AN - SCOPUS:85076710800
T3 - ACM International Conference Proceeding Series
BT - Proceedings of the Workshop on ACM RecSys Challenge, RecSys Challenge 2019
PB - Association for Computing Machinery
Y2 - 20 September 2019
ER -