TY - JOUR
KW - Collaborative Filtering
KW - Neural Networks
KW - One-Hot Encoding
KW - Recommender System
KW - Siamese Networks
KW - Similarity Measure
AU - Jesús Bobadilla
AU - Abraham Gutiérrez
AB - Improving current similarity measures in the collaborative filtering Recommender Systems is relevant, since it contributes to different applications such as to get better big data representations of users and items, to implement dynamic browsers able to navigate through data, and to explain recommendation results. Currently, there are
many statistically based similarity measures, some of them tailored to the extraordinarily sparse collaborative filtering scenario. Nevertheless, the hypothesis of the paper is that using neural networks, learnt similarity measures can be obtained that improve existing ones. To accomplish the task, the typical neural models cannot be used, and it is necessary to focus on the similarity learning area, in which the goal is to make the modellearn, which is a similarity function able to measure how similar two objects are. Siamese networks adequately implement the similarity learning concept, and we have adapted them to collaborative filtering particularities. The results in different scenarios show significant improvements compared to the state-of-the-art.
IS - In press
M1 - In press
N2 - Improving current similarity measures in the collaborative filtering Recommender Systems is relevant, since it contributes to different applications such as to get better big data representations of users and items, to implement dynamic browsers able to navigate through data, and to explain recommendation results. Currently, there are
many statistically based similarity measures, some of them tailored to the extraordinarily sparse collaborative filtering scenario. Nevertheless, the hypothesis of the paper is that using neural networks, learnt similarity measures can be obtained that improve existing ones. To accomplish the task, the typical neural models cannot be used, and it is necessary to focus on the similarity learning area, in which the goal is to make the modellearn, which is a similarity function able to measure how similar two objects are. Siamese networks adequately implement the similarity learning concept, and we have adapted them to collaborative filtering particularities. The results in different scenarios show significant improvements compared to the state-of-the-art.
PY - 9998
SE - 1
SP - 1
EP - 7
T2 - International Journal of Interactive Multimedia and Artificial Intelligence
TI - Recommender Systems: Learning Collaborative Filtering Similarity Measures Using Siamese Networks
UR - https://www.ijimai.org/journal/bibcite/reference/3573
VL - In press
SN - 1989-1660
ER -