Abstract:Multimedia content is dominating today's Web information. The nature of multimedia user-item interactions is 1/0 binary implicit feedback (e.g., photo likes, video views, song downloads, etc.), which can be collected at a larger scale with a much lower cost than explicit feedback (e.g., product ratings). However, the majority of existing collaborative filtering (CF) systems are not well-designed for multimedia recommendation, since they ignore the implicitness in users' interactions with multimedia content. We argue that, in multimedia recommendation, there exists item- and component-level implicitness which blurs the underlying users' preferences. The item-level implicitness means that users' preferences on items (e.g. photos, videos, songs, etc.) are unknown, while the component-level implicitness means that inside each item users' preferences on different components (e.g. regions in an image, frames of a video, etc.) are unknown. For example, a 'view'' on a video does not provide any specific information about how the user likes the video (i.e.item-level) and which parts of the video the user is interested in (i.e.component-level). In this paper, we introduce a novel attention mechanism in CF to address the challenging item- and component-level implicit feedback in multimedia recommendation, dubbed Attentive Collaborative Filtering (ACF). Specifically, our attention model is a neural network that consists of two attention modules: the component-level attention module, starting from any content feature extraction network (e.g. CNN for images/videos), which learns to select informative components of multimedia items, and the item-level attention module, which learns to score the item preferences. ACF can be seamlessly incorporated into classic CF models with implicit feedback, such as BPR and SVD++, and efficiently trained using SGD. Through extensive experiments on two real-world multimedia Web services: Vine and Pinterest, we show that ACF significantly outperforms state-of-the-art CF methods.

Implicit Rating Methods Based on Interest Preferences of Categories for Micro-Video Recommendation

Confidence-Learning Based Collaborative Filtering With Heterogeneous Implicit Feedbacks

Implicit semantic-based personalized micro-videos recommendation

Brand Recommendation Leveraging Heterogeneous Implicit Feedbacks

CMBPR: Category-Aided Multi-Channel Bayesian Personalized Ranking for Short Video Recommendation

Online Collaborative Filtering With Implicit Feedback

An Intelligent Video Tag Recommendation Method for Improving Video Popularity in Mobile Computing Environment

Improving Micro-video Recommendation via Contrastive Multiple Interests

Implicit Recommendation with Interest Change and User Influence

Learning Fine-grained User Interests for Micro-video Recommendation

Personalized Micro-Video Recommendation via Hierarchical User Interest Modeling

Attentive Collaborative Filtering: Multimedia Recommendation with Item- and Component-Level Attention

Preference Transfer Model in Collaborative Filtering for Implicit Data

Learning and Fusing Multiple User Interest Representations for Micro-Video and Movie Recommendations

Personalized Micro-video Recommendation Based on Multi-modal Features and User Interest Evolution

Real-time Video Recommendation Exploration

Explicit Feedbacks Meet with Implicit Feedbacks : A Combined Approach for Recommendation System

Learning and Optimization of Implicit Negative Feedback for Industrial Short-video Recommender System

MMM: Multi-source Multi-net Micro-video Recommendation with Clustered Hidden Item Representation Learning

Temporal Diversity-Aware Micro-Video Recommendation with Long- and Short-Term Interests Modeling