AMM: Attentive Multi-field Matching for News Recommendation

Personalized news recommendation methods are widely used in online news services. These methods usually recommend news based on the matching between news content and user interest inferred from historical behaviors. However, these methods usually have difficulties in making accurate recommendations to cold-start users, and tend to recommend similar news with those users have read. In general, popular news usually contain important information and can attract users with different interests. Besides, they are usually diverse in content and topic. Thus, in this paper we propose to incorporate news popularity information to alleviate the cold-start and diversity problems for personalized news recommendation. In our method, the ranking score for recommending a candidate news to a target user is the combination of a personalized matching score and a news popularity score. The former is used to capture the personalized user interest in news. The latter is used to measure time-aware popularity of candidate news, which is predicted based on news content, recency, and real-time CTR using a unified framework. Besides, we propose a popularity-aware user encoder to eliminate the popularity bias in user behaviors for accurate interest modeling. Experiments on two real-world datasets show our method can effectively improve the accuracy and diversity for news recommendation.

Personalized news recommendation is important for users to find interesting news information and alleviate information overload. Although it has been extensively studied over decades and has achieved notable success in improving user experience, there are still many problems and challenges that need to be further studied. To help researchers master the advances in personalized news recommendation, in this article, we present a comprehensive overview of personalized news recommendation. Instead of following the conventional taxonomy of news recommendation methods, in this article, we propose a novel perspective to understand personalized news recommendation based on its core problems and the associated techniques and challenges. We first review the techniques for tackling each core problem in a personalized news recommender system and the challenges they face. Next, we introduce the public datasets and evaluation methods for personalized news recommendation. We then discuss the key points on improving the responsibility of personalized news recommender systems. Finally, we raise several research directions that are worth investigating in the future. This article can provide up-to-date and comprehensive views on personalized news recommendation. We hope this article can facilitate research on personalized news recommendation as well as related fields in natural language processing and data mining.

METoNR: A meta explanation triplet oriented news recommendation model

News representation is critical for news recommendation. Most existing methods learn news representations only from news texts while ignoring the visual information of news. In fact, users may click news not only due to the interest in news titles but also the attraction of news images. Thus, images are useful for representing news and predicting news clicks. Pretrained visiolinguistic models are powerful in multi-modal understanding, which can represent news from both textual and visual contents. In this paper, we propose a multimodal news recommendation method that can incorporate both textual and visual information of news to learn multimodal news representations. We first extract region-of-interests (ROIs) from news images via object detection. We then use a pre-trained visiolinguistic model to encode both news texts and image ROIs and model their inherent relatedness using co-attentional Transformers. In addition, we propose a crossmodal candidate-aware attention network to select relevant historical clicked news for the accurate modeling of user interest in candidate news. Experiments validate that incorporating multimodal news information can effectively improve the performance of news recommendation.

A personalised news recommendation system extracts news set from multiple press releases and presents the recommended news to the user. In an effort to build a better recommender system with high accuracy, this paper proposes a personalised news recommendation framework named Hybrid Personalised NEws Recommendation (HYPNER). HYPNER combines both collaborative filtering-based and content-based filtering methods. The proposed framework aims at improving the accuracy of news recommendation by resolving the issues of scalability due to large news corpus, enriching the user's profile, representing the exact properties and characteristics of news items, and recommending diverse set of news items. Validation experiments showed that HYPNER achieved 81.56% improvement in F1 -score and 5.33% in diversity as compared to an existing recommender system, SCENE.

In news recommendation system, the content-based method face the lack of diversity that recommended results will only contain the news similar with user's former read. The hybrid method face cold-start problem that the candidate news must wait enough click to recommend. We propose a fused method which is capable of handling the diversity problem of content-based and cold-start problem of hybrid approach. We use content-based method to extract user's existed interest firstly. Find out similar user set and predict target user's potential interest by collaborative filtering secondly. Fuse existed interest and potential interest to obtain the final user profile the third. Generate the recommended results after comparison the similarity of candidate news and final user profile lastly. This approach results in an offline news set and compared Diversity and F-score with other recommendation method. It decreases the lack of diversity compare with content-based approach and avoids the cold-start problem compare with hybrid method.

Recommender systems are becoming an essential part of smart services. When building a news recommender system, we should consider special features different from other recommender systems. Hot news topics are changing every moment, thus it is important to recommend right news at the right time. This paper aims to propose a new model, based on deep neural network, to analyse user preference for news recommender system. The model extracts interest keywords to characterize the user preference from the set of news articles read by that particular user in the past. The model utilizes characterizing features for news recommendation, and applies those to the keyword classification for user preference. For the keyword classification, we use deep neural network for online preference analysis, because adaptive learning is necessary to track changes of hot topics sensitively. The usefulness of our model is validated through experiments. In addition, the accuracy and diversity of the recommendation results is also analysed.

Recommender systems are designed to help users quickly access large volumes of information according to their profiles. Most previous works in recommender systems have put their emphasis on the accuracy of finding the most similar items according to a user’s profile, while often ignoring other aspects that may affect users’ experiences in practice, e.g., the novelty and diversity issues within a recommendation list. In this paper, we focus on utilizing taxonomic knowledge extracted from an online encyclopedia to boost a content-based personalized news recommender system without much human involvement. Given a recommendation list, we improve a user’s satisfaction by introducing the taxonomy based novelty and diversity metrics to include novel, but potentially related items into the list, and filter out redundant ones. The experimental results show that the coarse grained knowledge resources can help a content-based news recommender system provides accurate as well as user-oriented recommendations.

In recent years, the type and quantity of news are growing rapidly, and it is not easy for users to find the news they are interested in the massive amount of news. A news recommendation system can score and predict the candidate news, and finally recommend the news with high scores to users. However, existing user models usually only consider users’ long-term interests and ignore users’ recent interests, which affects users’ usage experience. Therefore, this paper introduces gated recurrent unit (GRU) sequence network to capture users’ short-term interests and combines users’ short-term interests and long-term interests to characterize users. While existing models often only use the user’s browsing history and ignore the variability of different users’ interest in the same news, we introduce additional user’s ID information and apply the personalized attention mechanism for user representation. Thus, we achieve a more accurate user representation. We also consider the risk of compromising user privacy if the user model training is placed on the server side. To solve this problem, we design the training of the user model locally on the client side by introducing a federated learning framework to keep the user’s browsing history on the client side. We further employ secure multi-party computation to request news representations from the server side, which protects privacy to some extent. Extensive experiments on a real-world news dataset show that our proposed news recommendation model has a better improvement in several performance evaluation metrics. Compared with the current state-of-the-art federated news recommendation models, our model has increased by 0.54% in AUC, 1.97% in MRR, 2.59% in nDCG@5%, and 1.89% in nDCG@10. At the same time, because we use a federated learning framework, compared with other centralized news recommendation methods, we achieve privacy protection for users.

SI-News: Integrating social information for news recommendation with attention-based graph convolutional network

Using complex language models has been a common strategy as personalized news recommendation systems are adopted by online news sites more and more. Before the epoch of GPT-3, news recommendation systems underwent a progression from rule-based and collaborative filtering approaches in the pre-2010s, through the integration of neural networks in the 2010s, to the emergence of earlier iterations of large language models like GPT-2 in 2019. Pre-trained language models have ushered in a new era of recommendation paradigms, thanks to the emergence of huge language models like GPT-3 and T-5. With its easy-to-use interface, ChatGPT is becoming more and more popular for text-based jobs. Focusing on news provider fairness, individualized news recommendations, and fake news identification, this study starts an inquiry into ChatGPT's efficacy in news recommendations. We acknowledge that ChatGPT's output sensitivity to input phrasing is a limitation, and our goal is to investigate these limitations from each angle. We also investigate whether certain prompt formats can help to alleviate these constraints or if more research is needed. To go beyond set assessments, we create a webpage where ChatGPT's performance on the examined activities and prompts is tracked once a week. Through the use of big language models, this work seeks to improve news recommendation performance and stimulate more research in this area.

DuCL: Dual-stage contrastive learning framework for Chinese semantic textual matching

Recently, exploiting a knowledge graph (KG) to enrich the semantic representation of a news article have been proven to be effective for news recommendation. These solutions focus on the representation learning for news articles with additional information in the knowledge graph, where the user representations are mainly derived based on these news representations later. However, different users would hold different interests on the same news article. In other words, directly identifying the entities relevant to the user's interest and deriving the resultant user representation could enable a better news recommendation and explanation.

Online news are being spread through social media by news agencies to encourage people to read news from their site. After users log in to their site, users will continue to read the news if it is relevant to their personalized news recommendation. However, nowadays, a personalized recommendation could be provided to users if the site has records of users' browsing history, and users must have logged in to the news site. This could be problematic if the readers are anonymous users, and the site does not have enough browsing history records of these users. It will be difficult for news agencies to increase the number of daily readers on their site, especially if they have to compete with other agencies for providing a good personalized recommendation. Therefore, a personalized recommendation system is necessary for anonymous users who only visit the news site occasionally so that the website can recommend the news they prefer to read by implementing CSR Sparse Matrix Vector Multiplication and proximity processing as sparse matrix applied method. The Bayesian framework for user interest will also be implemented with the modification in its term. This research proved that personalized news recommendations could still be provided to readers with limited knowledge of the users' searching history.

Online news recommender systems aim to make personalized recommendations according to user preferences, which require modeling users’ short-term reading interest. However, due to the limited logged user interactions in practice, news recommendation at session-level becomes very challenging. Existing methods on session-based news recommendation mainly focus on extracting features from news articles and sequential user-item interactions, but they usually ignore the semantic-level structural information among news articles and do not explore external knowledge sources. In this paper, we propose a novel Context-Aware Graph Embedding (CAGE) framework for session-based news recommendation, which builds an auxiliary knowledge graph to enrich the semantic meaning of entities involved in articles, and further refines the article embeddings by graph convolutional networks. Experimental results on a real-world news dataset demonstrate the effectiveness of our method compared with the state-of-the-art methods on session-based news recommendation.