https://opus.lib.uts.edu.au/bitstream/10453/44133/1/JCSS_Qusai.pdf

An effective recommender system by unifying user and item trust information for B2B applications

Personalized Service System Based on Hybrid Filtering for Digital Library

Recent advances in path-based explainable recommendation systems have attracted increasing attention thanks to the rich information provided by knowledge graphs. Most existing explainable recommendation only utilizes static knowledge graph and ignores the dynamic user-item evolutions, leading to less convincing and inaccurate explanations. Although there are some works that realize that modelling user's temporal sequential behaviour could boost the performance and explainability of the recommender systems, most of them either only focus on modelling user's sequential interactions within a path or independently and separately of the recommendation mechanism. In this paper, we propose a novel Temporal Meta-path Guided Explainable Recommendation (TMER), which utilizes well-designed item-item path modelling between consecutive items with attention mechanisms to sequentially model dynamic user-item evolutions on dynamic knowledge graph for explainable recommendations. Compared with existing works that use heavy recurrent neural networks to model temporal information, we propose simple but effective neural networks to capture users' historical item features and path-based context to characterise next purchased item. Extensive evaluations of TMER on three real-world benchmark datasets show state-of-the-art performance compared against recent strong baselines.

Temporal Meta-path Guided Explainable Recommendation

With the development of the Internet, the problem of information overload is becoming increasing serious. People all have experienced the feeling of being overwhelmed by the number of new books, articles, and proceedings coming out each year. Many researchers pay more attention on building a proper tool which can help users obtain personalized resources. Personalized recommendation systems are one such software tool used to help users obtain recommendations for unseen items based on their preferences. The commonly used personalized recommendation system methods are content-based filtering, collaborative filtering, and association rules mining. Unfortunately, each method has its drawbacks. This paper presented a personalized collaborative filtering recommendation method combining the association rules mining and self-organizing map. It used the association rules mining to fill the vacant where necessary. Then, it employs clustering function of self-organizing map to form nearest neighbors of the target item and it produces prediction of the target user to the target item using item-based collaborative filtering. The recommendation method combining association rules mining and collaborative filtering can alleviate the data sparsity problem in the recommender systems.

/pdf/a-personalized-collaborative-filtering-recommendation-using-3pxv4qozsn.pdf

A Personalized Collaborative Filtering Recommendation Using Association Rules Mining and Self-Organizing Map

Product attribute values are essential in many e-commerce scenarios, such as customer service robots, product recommendations, and product retrieval. While in the real world, the attribute values of a product are usually incomplete and vary over time, which greatly hinders the practical applications. In this paper, we propose a multimodal method to jointly predict product attributes and extract values from textual product descriptions with the help of the product images. We argue that product attributes and values are highly correlated, e.g., it will be easier to extract the values on condition that the product attributes are given. Thus, we jointly model the attribute prediction and value extraction tasks from multiple aspects towards the interactions between attributes and values. Moreover, product images have distinct effects on our tasks for different product attributes and values. Thus, we selectively draw useful visual information from product images to enhance our model. We annotate a multimodal product attribute value dataset that contains 87,194 instances, and the experimental results on this dataset demonstrate that explicitly modeling the relationship between attributes and values facilitates our method to establish the correspondence between them, and selectively utilizing visual product information is necessary for the task. Our code and dataset are available at https://github.com/jd-aig/JAVE.

Multimodal Joint Attribute Prediction and Value Extraction for E-commerce Product

Recommender systems are web based systems that aim at predicting a customer's interest on available products and services by relying on previously rated products and dealing with the problem of information and product overload. Collaborative filtering is the most popular recommendation technique nowadays and it mainly employs the user item rating data set . Traditional collaborative filtering approaches compute a similarity value between the target user and each other user by computing the relativity of their ratin gs , which is the set of ratings given on the same items. Based on the ratings of the most similar users, commonly referred to as neighbors, the algorithms compute recommendations for the target user. They only consider the ratings information. User attribute information associated with a user's personality and item attribute information associated with a n item's inside are rarely considered in the collaborative filtering recommendation process. In this paper, a new collaborative filtering personalized recommendation algorithm is proposed which employs the user attribute information and the item attribute information . This approach combines the user rating similarity and the user attribute similarity in the user based collaborative filtering process to fill the vacant ratings where necessary, and then it combines the item rating similarity and the item attribute similarity in the item based collaborative filtering process to produce recommendations . The hybrid collaborative filtering employs the user attribute and item attribute can alleviate the sparsity issue in the recommender systems.

SongJie Gong

Papers

Employing User Attribute and Item Attribute to Enhance the Collaborative Filtering Recommendation