News reading is an important social activity and to help readers quickly find news articles of their interest, news content providers and aggregators use recommender systems. Such systems are ...designed to address a variety of challenges. Inspiration for algorithmic design is taken from various domains which has resulted in the creation of an enormous body of literature. Also, different methods are used for evaluation of the recommendation algorithms. In this study, we review these developments and present three major components in news recommendation research. First, we list and categorise the challenges faced while designing news recommender systems. We especially list the different algorithmic designs used for generating personalised and non-personalised recommendations. We discuss the major neural network architectures that are being increasingly used for both collaborative and content-based recommender systems. Next, we list the two major evaluation methods and also list some popular datasets used in evaluation. Finally, we identify the emerging trends in news recommender research. We find that the issues related to fake news, trust and use of personal data for news recommendation are gaining wider attention, and deep learning methods are being increasingly used to address these issues.
Abstract
Sports recommender systems receive an increasing attention due to their potential of fostering healthy living, improving personal well-being, and increasing performances in sports. These ...systems support people in sports, for example, by the recommendation of healthy and performance-boosting food items, the recommendation of training practices, talent and team recommendation, and the recommendation of specific tactics in competitions. With applications in the virtual world, for example, the recommendation of maps or opponents in e-sports, these systems already transcend conventional sports scenarios where physical presence is needed. On the basis of different examples, we present an overview of sports recommender systems applications and techniques. Overall, we analyze the related state-of-the-art and discuss future research directions.
Data transfer across numerous platforms has increased dramatically due to the enormous number of visitors or users of the present e-commerce platform. With the rise of increasingly massive data, ...consumers are finding it challenging to obtain the right data. The recommendation engine may be used to make it simpler to find information that is relevant to the user's needs. Clothing, gadgets, autos, furniture, and other e-commerce items rely on product visualization to entice shoppers. There are millions of images in these items. Displaying the information sought by clients based on visual data is a difficult challenge to address. One strategy that is simple to use in a recommendation system is content-based filtering. This approach will eventually make suggestions to consumers based on previously accessible goods or product descriptions. Content-based filtering works by searching for similarities based on the properties of a product item. User interactions with a product will be recorded and analyzed in order to recommend certain similarities to users. Text-based datasets are used in the majority of content-based filtering studies. In this study, however, we attempt to leverage a dataset received from Kaggle in the form of images of futsal shoes. Then, VGG16 architecture is used to extract the image dataset. The top 5 most relevant item rankings are generated by this recommendation method using cosine similarity. In addition, the NDCG (Normalized Discounted Cumulative Gain) approach is used to assess the results of the suggestions. The NDCG was evaluated in ten test scenarios, with an average NDCG value of 0.855, indicating that the system delivers a reasonable performance suggestion.
Movie recommendation systems help movie enthusiasts by suggesting movies to watch without the hassle of having to go through the time-consuming process of deciding from a large collection of movie ...streaming platforms that recommend movies and TV episodes. News organizations that suggest articles to readers, and online stores that suggest products to customers all benefit from these recommendation systems. The algorithms implemented in this research train their models on the MovieLens dataset and provide users with tailored movie recommendations. The study compares different machine learning algorithms, which include a Content-based model, item-item and user-user collaborative filtering (CF), Collaborative filtering with Singular Value Decomposition (SVD), K Nearest Neighbors, and Non-negative Factorization. The algorithms are evaluated using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) to measure their accuracy and performance. While the proposed system which is based on a collaborative approach using SVD determines the connection between various users and, depending on their ratings, recommends movies to others with similar tastes, subsequently allowing users to explore more. The proposed approach using collaborative filtering with SVD performs better with a minimal RMSE of 0. 880258 by giving accurate and appropriate recommendations to the user. The model is further evaluated using performance metrics like Precision, Recall, and f1 score. So, CF with the SVD recommendation model is chosen for implementation and is integrated into a web application that allows the platform users to rate and review the available digital content as well as allows them to restrict screen time using a parental control system. The results of the study in this paper are presented in the form of tables, graphs, and statistical analyses, and can be used to guide the development of new and improved recommendation algorithms.
Background: Selecting a restaurant in a diverse city like Bandung can be challenging. This study leverages Twitter data and local restaurant information to develop an advanced recommendation system ...to improve decision-making. Objective: The system integrates content-based filtering (CBF) with deep feedforward neural network (DFF) classification to enhance the accuracy and relevance of restaurant recommendations. Methods: Data was sourced from Twitter and PergiKuliner, with restaurant-related tweets converted into rating values. The CBF combined Bag of Words (BoW) and cosine similarity, followed by DFF classification. SMOTE was applied during training to address data imbalance. Results: The initial evaluation of CBF showed a Mean Absolute Error (MAE) of 0.0614 and a Root Mean Square Error (RMSE) of 0.0934. The optimal DFF configuration in the first phase used two layers with 32/16 nodes, a dropout rate of 0.3, and a 20% test size. This setup achieved an accuracy of 81.08%, precision of 82.89%, recall of 76.93%, and f1-scores of 79.23%. In the second phase, the RMSprop optimizer improved accuracy to 81.30%, and tuning the learning rate to 0.0596 further increased accuracy to 89%, marking a 9.77% improvement. Conclusion: The research successfully developed a robust recommendation system, significantly improving restaurant recommendation accuracy in Bandung. The 9.77% accuracy increase highlights the importance of hyperparameter tuning. SMOTE also proved crucial in balancing the dataset, contributing to a well-rounded learning model. Future studies could explore additional contextual factors and experiment with recurrent or convolutional neural networks to enhance performance further.
Web 2.0 platforms such as blogs, online news, social networks, and Internet forums allow users to write comments to express their interests and opinions about the content of news articles, videos, ...blogs or forum posts, etc. Users’ comments contain additional information about the content of Web documents as well as provide important means for user interactions. In this paper, we present a study on the task of recommending, for a given user, a short list of suitable stories for commenting. We propose an efficient collaborative filtering method which exploits co-commenting patterns of users to generate recommendations. To further improve the accuracy, we also introduce a novel hybrid recommendation method that combines the proposed collaborative features and content based features in a learning-to-rank framework. We verify the effectiveness of the proposed methods on two datasets including samples of user comments from an online forum and a forum-based news service. Experimental results show that the proposed collaborative filtering method substantially outperforms traditional content-based approaches in terms of accuracy. Furthermore, the proposed hybrid approach leads to additional improvements over individual recommendation methods and achieves higher accuracy than a baseline hybrid approach. The results also demonstrate the stability of our methods in handling newly posted stories with a small number of comments.
•A new recommender system is represented which has three parts.•Content-based, collaborative, and hybrid filtering are three sections of the proposed system.•This work is developed on discussion ...groups with tagging feature.•Semantic relevancies of tags are extracted using WordNet database.•The tags are organized in a hierarchical structure based on their semantic relevance.
Discussion groups are one of the most important elements of collaborative learning which utilize recommender systems to improve their performance in several aspects. This type of learning facilitates a comfort communication between users to share their problems and questions and receive the appropriate solutions. Most of recommender systems of discussion groups are based on using collaborative filtering techniques and a few numbers of them use content-based or hybrid filtering. Experimental results of previous works show that using hybrid recommender systems on discussion groups’ databases cause significant improvement in accuracy of recommended posts in comparison with other filtering techniques (Kardan and Ebrahimi, 2013). To improve performance of (Kardan and Ebrahimi, 2013), in this paper, a new recommender system is represented, which includes three parts, namely content-based, collaborative, and hybrid filtering parts. The proposed recommender system uses the tagging features to provide more appropriate recommendations on discussion groups. For this purpose, semantic relevance of tags is extracted using WordNet lexical database and the tags are organized in a hierarchical structure based on their semantic relevance. The hierarchical structure is used for searching relevant posts in content-based filtering part, and the user’s query is extended using related semantic tags. The implicit ratings of the users are calculated in the collaborative filtering part using similarity measures. Finally, the results of these two parts are combined in the hybrid filtering part of the proposed system to recommend the posts of the discussion group which are similar to the query of the active user. Experimental results show higher precision of the proposed system comparing to the former recommender systems.
Recommendation systems (RSs) have garnered immense interest for applications in e-commerce and digital media. Traditional approaches in RSs include such as collaborative filtering (CF) and ...content-based filtering (CBF) through these approaches that have certain limitations, such as the necessity of prior user history and habits for performing the task of recommendation. To minimize the effect of such limitation, this article proposes a hybrid RS for the movies that leverage the best of concepts used from CF and CBF along with sentiment analysis of tweets from microblogging sites. The purpose to use movie tweets is to understand the current trends, public sentiment, and user response of the movie. Experiments conducted on the public database have yielded promising results.
Linked Data allows structured data to be published in a standard manner so that datasets from diverse domains can be interlinked. By leveraging Semantic Web standards and technologies, a growing ...amount of semantic content has been published on the Web as Linked Open Data (LOD). The LOD cloud has made available a large volume of structured data in a range of domains via liberal licenses. The semantic content of LOD in conjunction with the advanced searching and querying mechanisms provided by SPARQL has opened up unprecedented opportunities not only for enhancing existing applications, but also for developing new and innovative semantic applications. However, SPARQL is inadequate to deal with functionalities such as comparing, prioritizing, and ranking search results which are fundamental to applications such as recommendation provision, matchmaking, social network analysis, visualization, and data clustering. This paper addresses this problem by developing a systematic measurement model of semantic similarity between resources in Linked Data. By drawing extensively on a feature-based definition of Linked Data, it proposes a generalized information content-based approach that improves on previous methods which are typically restricted to specific knowledge representation models and less relevant in the context of Linked Data. It is validated and evaluated for measuring item similarity in recommender systems. The experimental evaluation of the proposed measure shows that our approach can outperform comparable recommender systems that use conventional similarity measures.
PDF
