User-based Collaborative Filtering Approach Research Articles

A hybrid recommendation algorithm for green food based on review text and review time

ABSTRACT Green food is well-known for its health benefits, environmental-friendliness, and safety. Current recommender systems used by e-commerce websites usually recommend products based on products' popularity or customers' ratings. However, users' reviews could be more representative of consumers' preferences. In addition, users' review time is not utilized. To reduce the recommendation bias, this study proposes a hybrid recommendation algorithm based on green food reviews and review time. The proposed algorithm combines a content-based recommendation algorithm with a user-based collaborative filtering approach, where affective values of reviews replace ratings and a time impact factor is considered. With the two classical evaluation indices of F1 and Mean Absolute Error (MAE), the experiments show that considering both reviews’ sentiments and dynamic changes of individuals’ preferences could improve recommendation effectiveness over three other algorithms, which provides a new reference direction for improving existing recommender systems on green food.

Drug Recommendation System for Diabetes Using a Collaborative Filtering and Clustering Approach: Development and Performance Evaluation.

BackgroundDiabetes is a public health problem worldwide. Although diabetes is a chronic and incurable disease, measures and treatments can be taken to control it and keep the patient stable. Diabetes has been the subject of extensive research, ranging from disease prevention to the use of technologies for its diagnosis and control. Health institutions obtain information required for the diagnosis of diabetes through various tests, and appropriate treatment is provided according to the diagnosis. These institutions have databases with large volumes of information that can be analyzed and used in different applications such as pattern discovery and outcome prediction, which can help health personnel in making decisions about treatments or determining the appropriate prescriptions for diabetes management.ObjectiveThe aim of this study was to develop a drug recommendation system for patients with diabetes based on collaborative filtering and clustering techniques as a complement to the treatments given by the treating doctor.MethodsThe data set used contains information from patients with diabetes available in the University of California Irvine Machine Learning Repository. Data mining techniques were applied for processing and analysis of the data set. Unsupervised learning techniques were used for dimensionality reduction and patient clustering. Drug predictions were obtained with a user-based collaborative filtering approach, which enabled creating a patient profile that can be compared with the profiles of other patients with similar characteristics. Finally, recommendations were made considering the identified patient groups. The performance of the system was evaluated using metrics to assess the quality of the groups and the quality of the predictions and recommendations.ResultsPrincipal component analysis to reduce the dimensionality of the data showed that eight components best explained the variability of the data. We identified six groups of patients using the clustering algorithm, which were evenly distributed. These groups were identified based on the available information of patients with diabetes, and then the variation between groups was examined to predict a suitable medication for a target patient. The recommender system achieved good results in the quality of predictions with a mean squared error metric of 0.51 and accuracy in the quality of recommendations of 0.61, which is acceptable.ConclusionsThis work presents a recommendation system that suggests medications according to drug information and the characteristics of patients with diabetes. Some aspects related to this disease were analyzed based on the data set used from patients with diabetes. The experimental results with clustering and prediction techniques were found to be acceptable for the recommendation process. This system can provide a novel perspective for health institutions that require technologies to support health care personnel in the management of diabetes treatment and control.

Using tags for measuring the semantic similarity of users to enhance collaborative filtering recommender systems

Recent years have seen a significant growth in social tagging systems, which allow users to use their own generated tags to organize, categorize, describe and search digital content on social media. The growing popularity of tagging systems is leading to an increasing need for automatic generation of recommended items for users. Much previous research focuses on incorporating recommender techniques in social tagging systems to support the suggestion of suitable tags for annotating related items. Collaborative filtering is one such technique. The most critical task in collaborative filtering is finding related users with similar preferences, i.e., “liked-minded” users. Despite the popularity of collaborative filtering, it still suffers from certain limitations in relation to “cold-start” users, for example, where often there are insufficient preferences to make recommendations. Moreover, there is the data-sparsity problem, where there is limited user feedback data to identify similarities in users’ interests because there is no intersection between users’ transactional data a situation which also results in degraded recommendation quality. For this reason, in this paper we present a new collaborative filtering approach based on users’ semantic tags, which calculates the similarity between users by discovering the semantic spaces in their posted tags. We believe that this approach better reflects the semantic similarity between users according to their tagging perspectives and consequently improves recommendations through the identification of semantically related items for each user. Our experiment on a real-life dataset shows that the proposed approach outperforms the traditional user-based collaborative filtering approach in terms of improving the quality of recommendations.

User based Recommender Systems using Implicative Rating Measure

This paper proposes the implicative rating measure developed on the typicality measure. The paper also proposes a new recommendation model presenting the top N items to the active users. The proposed model is based on the user-based collaborative filtering approach using the implicative intensity measure to find the nearest neighbors of the active users, and the proposed measure to predict users’ ratings for items. The model is evaluated on two datasets MovieLens and CourseRegistration, and compared to some existing models such as: the item based collaborative filtering model using the Jaccard measure, the user based collaborative filtering model using Jaccard measure, the popular items based model, the latent factor based model, and the association rule based model using the confidence measure. The experimental results show that the performance of the proposed model is better when compared to other five models.

Towards Trajectory-Based Recommendations in Museums: Evaluation of Strategies Using Mixed Synthetic and Real Data

Recommendation systems, which suggest items that are of potential interest to the user (e.g., regarding which books to read, which movies to watch, etc.) have grown in popularity due to the ever-increasing amount of data available, that can lead to significant user’s overload. In particular, in recent years, extensive research has focused on the so-called Context-Aware Recommender Systems (CARS), which exploit context data to offer more relevant recommendations.In this paper, we study this problem with a use case scenario: recommending items to observe in a museum. We propose a trajectory-based and user-based collaborative filtering approach, that considers context data such as the location of the user and his/her trajectory to offer personalized recommendations. Besides, we exploit DataGenCARS, a dataset synthetic generator designed to construct datasets for the evaluation of context-aware recommendation systems, to build a mixed scenario based on both real and synthetic data. The experimental results show the advantages of the proposed approach and the usefulness of DataGenCARS for practical evaluation with a real use-case scenario.

Context-aware location recommendation by using a random walk-based approach

The location-based social networks (LBSN) enable users to check in their current location and share it with other users. The accumulated check-in data can be employed for the benefit of users by providing personalized recommendations. In this paper, we propose a context-aware location recommendation system for LBSNs using a random walk approach. Our proposed approach considers the current context (i.e., current social relations, personal preferences and current location) of the user to provide personalized recommendations. We build a graph model of LBSNs for performing a random walk approach with restart. Random walk is performed to calculate the recommendation probabilities of the nodes. A list of locations are recommended to users after ordering the nodes according to the estimated probabilities. We compare our algorithm, CLoRW, with popularity-based, friend-based and expert-based baselines, user-based collaborative filtering approach and a similar work in the literature. According to experimental results, our algorithm outperforms these approaches in all of the test cases.

Hybrid content filtering and reputation-based popularity for recommending blog articles

Purpose – Social bookmarking is a system which allows users to share, organise, search and manage bookmarks of web resources. However, with the rapid growth in the production of online documents, people are facing the problem of information overload. Social bookmarking web sites offer a solution to this by providing push counts, which are counts of users’ recommendations of articles, and thus indicate the popularity and interest thereof. In this way, users can use the push counts to find popular and interesting articles. A measure of popularity-based solely on push counts, however, cannot be considered a true reflection of popularity. The paper aims to discuss these issues. Design/methodology/approach – In this paper, the authors propose to derive the degree of popularity of an article by considering the reputation of the users who push the article. Moreover, the authors propose a novel personalised blog article recommendation approach which combines reputation-based group popularity with content-based filtering (CBF), for the recommendation of popular blog articles which satisfy users’ personal preferences. Findings – The experimental results show that the proposed approach outperforms conventional CBF, item-based and user-based collaborative filtering approaches. The proposed approach considering reputation-based group popularity scores on neighbouring articles indeed can improve the recommendation quality of traditional CBF method. Originality/value – The recommendation approach modifies CBF method by considering the target user's group preferences, to overcome the limitation of CBF which arises from the recommending only items similar to those the user has previously liked. Users with similar article preferences (profiles) may form a group of users with similar interests. A group's preferences may also reflect an individual's preferences. The reputation-based group preferences of the target user's group can be used to complement the target user's preferences.

A Collaborative Filtering Recommendation Algorithm Based on Influence Sets

The traditional user-based collaborative filtering (CF) algorithms often suffer from two important problems: Scalability and sparsity because of its memory-based k nearest neighbor query algorithm. Item-Based CF algorithms have been designed to deal with the scalability problems associated with user-based CF approaches without sacrificing recommendation or prediction accuracy. However, item-based CF algorithms still suffer from the data sparsity problems. This paper presents a CF recommendation algorithm, named CFBIS (collaborative filtering based on influence sets), which is based on the concept of influence set and is a hot topic in information retrieval system. Moreover, it defines a new prediction computation method for this new recommendation mechanism. Experimental results show that the algorithm can achieve better prediction accuracy than traditional item-based CF algorithms. Furthermore, the algorithm can alleviate the dataset sparsity problem.