An Enhanced Multi-Modal Recommendation Based on Alternate Training With Knowledge Graph Representation

Abstract

Deep network recommendation is a cutting-edge topic in current recommendation system research, which as a combination of recommendation systems and deep learning theory can effectively improve recommendation accuracy. In a real recommendation scenario, all the effective information in a data set should be extracted, both explicit and implicit, because the comprehensive degree of information is proportional to the recommendation performance. This paper proposes an enhanced multi-modal recommendation based on alternate training with knowledge graph representation (SI-MKR) based on the MKR deep learning recommendation model. Our framework is an enhanced recommendation system based on knowledge graph representation, using valuable external knowledge as multi-modal information. The SI-MKR model solves the problem of ignoring the diversity of data types in the multi-modal knowledge-based recommendation system, which adds user and item attribute information from a knowledge graph as an enhancement recommendation multi-tasking training. By analysing the content of the item and user attributes, the SI-MKR model classifies the attributes of the items and users, processes the text type attributes and multi-value type attributes separately for feature extraction, and other types of attributes are used as inputs to the knowledge graph embedding unit. In addition, the knowledge graph data form a triplet unit, thus continuing the knowledge graph data training process. The feature extraction unit of the knowledge graph and the recommended unit are connected through the cross-compression unit for alternate training. During the deep learning framework training process, the recommendation system's item has a potential correlation with the head entity in the knowledge graph which embodies the idea of multi-tasking. Through extensive experiments on real-world datasets, we demonstrate that SI-MKR achieves substantial gains in movie recommendation over advanced model baselines. Even user-item interactions are sparse, SI-MKR maintains better performance than the MKR model.

Cinematic Curator: A Machine Learning Approach to Personalized Movie Recommendations

“…Future research attempts to separate noise from interactions that appear insignificant in order to uncover latent user preferences. Y. Wang, L. Dong, H. Zhang, X. Ma, Y. Li and M. Sun [3]. This paper presents SI-MKR, a sophisticated recommendation system that expands on the MKR deep learning model.…”

Multimodal Recommender Systems: A Survey

“…The multimodal features are fed into different modality encoders. The modality encoders extract the representations and are general architectures used in other fields, such as ViT [13] for images and General [34] Coarse-grained Attention CL [40] Coarse-grained Attention None [6], [21] Fine-gained Attention None [30], [27], [57] Combined Attention None [44], [39] User-item Graph + Fine-gained Attention None [56] User-item Graph CL [59] Item-item Graph CL [58], [38] Item-item Graph None [33] Item-item Graph + Fine-gained Attention None [50], [45] Knowledge Graph None [2], [46] Knowledge Graph CL [8] Knowledge Graph + Fine-gained Attention None [43] Knowledge Graph + Filtration (graph) None [63], [55], [31] Filtration (graph) None [49], [4] MLP / Concat DRL [15], [28] Fine-gained Attention DRL [61], [36], [48] None DRL…”

“…Furthermore, a novel graph attention network is adopted to aggregate neighboring entities while considering the relations in the knowledge graph. SI-MKR[50] proposes an enhanced multimodal recommendation method based on alternate training and the knowledge graph representation based on MKR[45]. Besides, most multimodal recommender systems ignore the problem of data type diversity.…”

Research on Learning Resource Recommendation Based on Knowledge Graph and Collaborative Filtering

“…The foundational TransE model, pioneered by Bordes et al [38], involves vectorizing entities and relationships in a vector space, aiming to make the vector addition of the head entity 'h' and the relationship 'r' approximately equal to the vector of the tail entity 't'. However, the TransE model exhibits limitations and is better suited for one-to-one relationships, struggling to perform well in scenarios involving one-to-many, many-to-one, or many-to-many relationships [39]. In response, subsequent models like TransH [40], TransR [41], and TransD [42] were introduced to address these shortcomings.…”