[PDF][PDF] On measuring popularity bias in collaborative filtering data
R Borges, K Stefanidis - 2020 - trepo.tuni.fi
2020•trepo.tuni.fi
The effect of having few data items responsible for the majority of ratings in a Collaborative
Filtering recommendation, and the complement of having majority of items responsible for
few ratings given by the users, are usually referred as popularity bias. The effect is known as
reflecting the preference of users for popular items, but also as a consequence of methods
and metrics normally applied by these systems. Variational Autoencoders (VAE) are
considered today the state-of-the-art for collaborative filtering recommenders, and can …
Filtering recommendation, and the complement of having majority of items responsible for
few ratings given by the users, are usually referred as popularity bias. The effect is known as
reflecting the preference of users for popular items, but also as a consequence of methods
and metrics normally applied by these systems. Variational Autoencoders (VAE) are
considered today the state-of-the-art for collaborative filtering recommenders, and can …
Abstract
The effect of having few data items responsible for the majority of ratings in a Collaborative Filtering recommendation, and the complement of having majority of items responsible for few ratings given by the users, are usually referred as popularity bias. The effect is known as reflecting the preference of users for popular items, but also as a consequence of methods and metrics normally applied by these systems. Variational Autoencoders (VAE) are considered today the state-of-the-art for collaborative filtering recommenders, and can handle big and sparse data entries with robustness and high accuracy. A methodology is proposed here for characterizing the popularity bias in Movielens and Netflix datasets, and when applying VAE for generating recommendations based on them. As a first step, the long tail model is applied for segmenting items and users in three different classes (Short Head, Medium Tail and Long Tail), depending on the proportion of interactions they are associated with. In addition, a real recommendation scenario is presented for measuring the proportion of unpopular items appearing among the suggestions provided by VAE. We consider characterizing the popularity in details as a very first step for providing recommenders with the desired serendipity effect, and expanding the knowledge of these systems about new and unpopular items with few ratings.
trepo.tuni.fi
以上显示的是最相近的搜索结果。 查看全部搜索结果