INTERPOS: Interaction Rhythm Guided Positional Morphing for Mobile App Recommender Systems

2025-06-14

AI Summary

To address the long temporal gaps between user interactions in mobile app recommendation—where conventional sequential models overlook users’ behavioral rhythms—this paper proposes a rhythm-guided positional deformation strategy. It explicitly models fine-grained rhythm signals as the actual time intervals (e.g., 1 day, 1 week, 1 month) between consecutive interactions and integrates them into positional encoding, thereby redefining the semantic meaning of positional embeddings in sequence representations. Built upon a Transformer-based autoregressive architecture, the method introduces rhythm-aware positional deformation embeddings and systematically explores three embedding fusion mechanisms: additive, multiplicative, and gated. Evaluated on seven real-world mobile app datasets, the model achieves average improvements of 3.2%–5.8% in NDCG@K and HIT@K over state-of-the-art methods. The implementation is publicly available.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

The mobile app market has expanded exponentially, offering millions of apps with diverse functionalities, yet research in mobile app recommendation remains limited. Traditional sequential recommender systems utilize the order of items in users' historical interactions to predict the next item for the users. Position embeddings, well-established in transformer-based architectures for natural language processing tasks, effectively distinguish token positions in sequences. In sequential recommendation systems, position embeddings can capture the order of items in a user's historical interaction sequence. Nevertheless, this ordering does not consider the time elapsed between two interactions of the same user (e.g., 1 day, 1 week, 1 month), referred to as"user rhythm". In mobile app recommendation datasets, the time between consecutive user interactions is notably longer compared to other domains like movies, posing significant challenges for sequential recommender systems. To address this phenomenon in the mobile app domain, we introduce INTERPOS, an Interaction Rhythm Guided Positional Morphing strategy for autoregressive mobile app recommender systems. INTERPOS incorporates rhythm-guided position embeddings, providing a more comprehensive representation that considers both the sequential order of interactions and the temporal gaps between them. This approach enables a deep understanding of users' rhythms at a fine-grained level, capturing the intricacies of their interaction patterns over time. We propose three strategies to incorporate the morphed positional embeddings in two transformer-based sequential recommendation system architectures. Our extensive evaluations show that INTERPOS outperforms state-of-the-art models using 7 mobile app recommendation datasets on NDCG@K and HIT@K metrics. The source code of INTERPOS is available at https://github.com/dlgrad/INTERPOS.

Problem

• Addresses limited research in mobile app recommendation systems
• Incorporates user rhythm in sequential recommendation models
• Improves accuracy in long-interval interaction sequences

Innovation

• Rhythm-guided position embeddings for sequences
• Morphed positional embeddings in transformers
• Fine-grained user interaction rhythm capture