Narrow Results

The creation of digital marketing has enabled companies to adopt personalized item recommendations for their customers. This process keeps them ahead of the competition. One of the techniques used in item recommendation is known as item-based recommendation system or item–item collaborative filtering. Presently, item recommendation is based completely on ratings like 1–5, which is not included in the comment section. In this context, users or customers express their feelings and thoughts about products or services. This paper proposes a machine learning model system where 0, 2, 4 are used to rate products. 0 is negative, 2 is neutral, 4 is positive. This will be in addition to the existing review system that takes care of the users’ reviews and comments, without disrupting it. We have implemented this model by using Keras, Pandas and Sci-kit Learning libraries to run the internal work. The proposed approach improved prediction with $79\%$ accuracy for Yelp datasets of businesses across 11 metropolitan areas in four countries, along with a mean absolute error (MAE) of $21\%$, precision at $79\%$, recall at $80\%$ and F1-Score at $79\%$. Our model shows scalability advantage and how organizations can revolutionize their recommender systems to attract possible customers and increase patronage. Also, the proposed similarity algorithm was compared to conventional algorithms to estimate its performance and accuracy in terms of its root mean square error (RMSE), precision and recall. Results of this experiment indicate that the similarity recommendation algorithm performs better than the conventional algorithm and enhances recommendation accuracy.

In recommender systems, Collaborative Filtering (CF) plays an essential role in promoting recommendation services. The conventional CF approach has limitations, namely data sparsity and cold-start. The matrix decomposition approach is demonstrated to be one of the effective approaches used in developing recommendation systems. This paper presents a new approach that uses CF and Singular Value Decomposition (SVD)$++$ for implementing a recommendation system. Therefore, this work is an attempt to extend the existing recommendation systems by (i) finding similarity between user and item from rating matrices using cosine similarity; (ii) predicting missing ratings using a matrix decomposition approach, and (iii) recommending top-N user-preferred items. The recommender system’s performance is evaluated considering Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). Performance evaluation is accomplished by comparing the systems developed using CF in combination with six different algorithms, namely SVD, SVD$++$, Co-Clustering, KNNBasic, KNNBaseline, and KNNWithMeans. We have experimented using MovieLens 100K, MovieLens 1M, and BookCrossing datasets. The results prove that the proposed approach gives a lesser error rate when cross-validation ($\mathrm{\text{CV}}=\{5,10,15\}$) is performed. The experimental results show that the lowest error rate is achieved with MovieLens 100K dataset ($\mathrm{\text{RMSE}}=0.9123$, $\mathrm{\text{MAE}}=0.7149$). The proposed approach also alleviates the sparsity and cold-start problems and recommends the relevant items.

Users’ ratings in recommender systems can be predicted by their historical data, item content, or preferences. In recent literature, scientists have used complex networks to model a user–user or an item–item network of the RS. Also, community detection methods can cluster users or items to improve the prediction accuracy further. However, the number of links in modeling a network is too large to do proper clustering, and community clustering is an NP-hard problem with high computation complexity. Thus, we combine fuzzy link importance and K-core decomposition in complex network models to provide more accurate rating predictions while reducing the computational complexity. The experimental results show that the proposed method can improve the prediction accuracy by 4.64$\%$ to 5.71$\%$ on the MovieLens data set and avoid solving NP-hard problems in community detection compared with existing methods. Our research reveals that the links in a modeled network can be reasonably managed by defining fuzzy link importance, and that the K-core decomposition can provide a simple clustering method with relatively low computation complexity.

Recommender systems have already been engaging multiple criteria for the production of recommendations. Such systems, referred to as multicriteria recommenders, demonstrated early the potential of applying Multi-Criteria Decision Making (MCDM) methods to facilitate recommendation in numerous application domains. On the other hand, systematic implementation and testing of multicriteria recommender systems in the context of real-life applications still remains rather limited. Previous studies dealing with the evaluation of recommender systems have outlined the importance of carrying out careful testing and parameterization of a recommender system, before it is actually deployed in a real setting. In this paper, the experimental analysis of several design options for three proposed multiattribute utility collaborative filtering algorithms is presented for a particular application context (recommendation of e-markets to online customers), under conditions similar to the ones expected during actual operation. The results of this study indicate that the performance of recommendation algorithms depends on the characteristics of the application context, as these are reflected on the properties of evaluations' data set. Therefore, it is judged important to experimentally analyze various design choices for multicriteria recommender systems, before their actual deployment.

Mobile recommender systems target on recommending the right product or information to the right mobile users at anytime and anywhere. It is well known that the contextual information is often the key for the performances of mobile recommendations. Therefore, in this paper, we provide a focused survey of the recent development of context-aware mobile recommendations. After briefly reviewing the state-of-the-art of recommender systems, we first discuss the general notion of mobile context and how the contextual information is collected. Then, we introduce the existing approaches to exploit contextual information for modeling mobile recommendations. Furthermore, we summarize several existing recommendation tasks in the mobile scenarios, such as the recommendations in the tourism domain. Finally, we discuss some key issues that are still critical in the field of context-aware mobile recommendations, including the privacy problem, the energy efficiency issues, and the design of user interfaces.

Nowadays, many online users find the selection of information and required products challenging due to the growing volume of data on the web. Recommender systems are introduced to deal with information overload. Cold start and data sparsity are the two primary issues in these systems, which lead to a decrease in the efficiency of recommender systems. To solve the problems, this paper proposes a novel method based on social network analysis. Our method leverages a multi-agent system for clustering users and items and predicting relationships between them simultaneously. The information on users and items is extracted from the user-item matrix as distinct graphs. Each of the graphs is then treated as a social network, which is further processed and analyzed by community detection and link prediction procedures. The users are grouped into several clusters by the community detection agent, which results in each cluster as a community. Then link prediction agent identifies the latent relationships between users and items. Simulation results show that the proposed method has significantly improved performance metrics as compared to recent techniques.

Matrix factorization models often reveal the low-dimensional latent structure in high-dimensional spaces while bringing space efficiency to large-scale collaborative filtering problems. Improving training and prediction time efficiencies of these models are also important since an accurate model may raise practical concerns if it is slow to capture the changing dynamics of the system. For the training task, powerful improvements have been proposed especially using SGD, ALS, and their parallel versions. In this paper, we focus on the prediction task and combine matrix factorization with approximate nearest neighbor search methods to improve the efficiency of top-N prediction queries. Our efforts result in a meta-algorithm, MMFNN, which can employ various common matrix factorization models, drastically improve their prediction efficiency, and still perform comparably to standard prediction approaches or sometimes even better in terms of predictive power. Using various batch, online, and incremental matrix factorization models, we present detailed empirical analysis results on many large implicit feedback datasets from different application domains.

Knowledge Graphs (KGs) have been shown to have great potential to provide rich and highly defined structured data about Recommender Systems (RSs) items. This paper introduces Explain- KGCN, an Explainable RS based on KGs and Graph Convolutional Networks (GCNs). The system emphasises the importance of semantic information characterisation and high-order connectivity of message passing to explore potential user preferences. Thus, based on a relation-specific neighbourhood aggregation function, it aims to generate for each given item a set of relation-specific embeddings that depend on each semantic relation in the KG. Specifically, the relation-specific aggregator discriminates neighbours based on their relationship with the target node, allowing the system to model the semantics of various relationships explicitly. Experiments conducted on two real-world datasets for the top-K recommendation task demonstrate the state-of-the-art performance of the system proposed. Besides improving predictive performance in terms of precision and recall, Explain-KGCN fully exploits wealthy structured information provided by KGs to offer recommendation explanation.

Recommender systems help users find relevant items efficiently based on their interests and historical interactions with other users. They are beneficial to businesses by promoting the sale of products and to user by reducing the search burden. Recommender systems can be developed by employing different approaches, including collaborative filtering (CF), demographic filtering (DF), content-based filtering (CBF) and knowledge-based filtering (KBF). However, large amounts of data can produce recommendations that are limited in accuracy because of diversity and sparsity issues. In this paper, we propose a novel hybrid method that combines user–user CF with the attributes of DF to indicate the nearest users, and compare four classifiers against each other. This method has been developed through an investigation of ways to reduce the errors in rating predictions based on users’ past interactions, which leads to improved prediction accuracy in all four classification algorithms. We applied a feature combination method that improves the prediction accuracy and to test our approach, we ran an offline evaluation using the 1M MovieLens dataset, well-known evaluation metrics and comparisons between methods with the results validating our proposed method.

Aiming to recommend potential collaborators for academic entities such as researchers and institutions, this paper develops a social recommender system through bibliometric indicators and network analytics. Targeting to scholarly articles, the proposed recommender system exploits co-authorships as established social relations and proposes a link prediction model for discovering such potential relations in terms of a co-authorship network. A case study recommending scientific collaborators for research entities on generelated diseases demonstrates the reliability of this study.

Network diffusion processes play an important role in solving the information overload problem. It has been shown that the diffusion-based recommendation methods have the advantage to generate both accurate and diverse recommendation items for online users. Despite that, numerous existing works consider the rating information as link weight or threshold to retain the useful links, few studies use the rating information to evaluate the recommendation results. In this paper, we measure the average rating of the recommended products, finding that diffusion-based recommendation methods have the risk of recommending low-rated products to users. In addition, we use the rating information to improve the network-based recommendation algorithms. The idea is to aggregate the diffusion results on multiple user-item bipartite networks each of which contains only links of certain ratings. By tuning the parameters, we find that the new method can sacrifice slightly the recommendation accuracy for improving the average rating of the recommended products.

With the rapid growth of commerce and development of Internet technology, a large number of user consumption preferences become available for online market intelligence analysis. A critical demand is to reduce the impact of information overload by using recommendation algorithms. In physical dynamics, network-based recommendation algorithms based on mass-diffusion have been popular for its simplicity and efficiency. In this paper, to solve the problem that most network-based recommendation algorithms cannot distinguish how much the user likes collected items and make resource configuration more reasonable, we propose a novel method called biased network-based inference (BNBI). The proposed method treats rating systems and nonrating systems differently and measures user’s preference for items by means of item similarity. The proposed method is evaluated in real datasets (MovieLens and Last.FM) and compared with some existing classic recommendation algorithms. Experimental results show that the proposed method is more effective and it can reduce the impact of item diversity and discover the real interest of users.

Clustering-based recommender systems bound the seek of similar users within small user clusters providing fast recommendations in large-scale datasets. Then groups can naturally be distributed into different data partitions scaling up in the number of users the recommender system can handle. Unfortunately, while the number of users and items included in a cluster solution increases, the performance in terms of precision of a clustering-based recommender system decreases. We present a novel approach that introduces a cluster-based distance function used for neighborhood computation. In our approach, clusters generated from the training data provide the basis for neighborhood selection. Then, to expand the search of relevant users, we use a novel measure that can exploit the global cluster structure to infer cluster-outside user’s distances. Empirical studies on five widely known benchmark datasets show that our proposal is very competitive in terms of precision, recall, and NDCG. However, the strongest point of our method relies on scalability, reaching speedups of 20× in a sequential computing evaluation framework and up to 100× in a parallel architecture. These results show that an efficient implementation of our cluster-based CF method can handle very large datasets providing also good results in terms of precision, avoiding the high computational costs involved in the application of more sophisticated techniques.

In the e-commerce arena new methods and tools have been recently developed to improve and customize the e-commerce web sites, according to users' necessities and preferences, that are usually vague and uncertain. The most successful tool in this field has been the Recommender Systems. Their aim is to assist e-shops customers to find out the most suitable products by using recommendations. Sometimes, these systems face situations where there is a lack of information or the information is vague or imprecise that yield unsuccessful results. Although several solutions have been proposed, they still present some limitations. In this paper, we present a Knowledge-Based Recommender System that manages and models the uncertainty related to users' preferences by using linguistic information. This system will overcome the problem of lack of information by computing recommendations through completing incomplete linguistic preference relations provided by the users.

With recent advances in e-commerce platforms, the information overload has grown due to increasing number of users, rapid generation of data and items in the recommender system. This tends to create serious problems in such recommender systems. The increasing features in recommender systems pose some new challenges due to poor resilience to mitigate against vulnerable attacks. In particular, the recommender systems are more prone to be attacked by shilling attacks, which creates more vulnerability. A recommender system with poor detection of attacks leads to a reduced detection rate. The performance of the recommender system is thus affected with poor detection ability. Hence, in this paper, we improve the resilience against shilling attacks using a modified Support Vector Machine (SVM) and a machine learning algorithm. The Gaussian Mixture Model is used as a machine learning algorithm to increase the detection rate and it further reduces the dimensionality of data in recommender systems. The proposed method is evaluated against several result metrics, such as the recall rate, precision rate and false positive rate between different attacks. The results of the proposed system are evaluated against probabilistic recommender approaches to demonstrate the efficacy of machine learning language in recommender systems.

Recommender systems nowadays are playing an important role in the delivery of services and information to users. Sentiment analysis (also known as opinion mining) is the process of determining the attitude of textual opinions, whether they are positive, negative or neutral. Data sparsity is representing a big issue for recommender systems because of the insufficiency of user rating or absence of data about users or items. This research proposed a hybrid approach combining sentiment analysis and recommender systems to tackle the problem of data sparsity problems by predicting the rating of products from users’ reviews using text mining and NLP techniques. This research focuses especially on Arabic reviews, where the model is evaluated using Opinion Corpus for Arabic (OCA) dataset. Our system was efficient, and it showed a good accuracy of nearly 85% in predicting the rating from reviews.

In the fast pace of life, E-learning has become a new way for self-improvement and competitiveness. The recommendation is needed in an E-learning system to filter suitable courses for users when they are facing a massive amount of information in course enrolment. However, due to the complexity of each learning course and the change of user interest, it is challenging to provide accurate recommendations. This paper proposes an E-learning recommender system that combines the recurrent neural network (RNN) and content-based technique to support users in course selection. The content-based techniques are to mine the relationships between courses, and the recurrent neural network is to extract user interests with a series of his/her enrolled courses. The proposed E-learning recommender system framework takes sequential connections into consideration. It intends to provide students with more precise course recommendations. The system is implemented with the Django framework and ElephantSQL cloud database and deployed on the Amazon Elastic Compute Cloud.

We describe a rating inference approach to incorporating textual user reviews into collaborative filtering (CF) algorithms. It elicits user preferences expressed in textual reviews, a problem known as sentiment analysis, and maps such preferences onto rating scales that can be understood by existing CF algorithms. Our rating inference approach identifies opinion words from user reviews, and then estimates the sentimental orientations (SO) and strengths of the opinion words based on their relative frequencies of occurrence in reviews having different sentiments. This method allows similar words to have different SO, thereby addresses a major limitation of existing methods. A rating is assigned to a review based on the average SO of the opinion words it contains. An empirical evaluation was carried out to study the effects of various factors on the proposed framework. Empirical results validated the effectiveness of the framework, and suggest that it does not rely on a large training corpus.

Nowadays, the Recommender System (RS) has gained much attention in industry and academia. RS offers valuable ideas to users while interacting with a website or application. With the advancements in the mobile environment (ME), the users’ interest has turned to the movie recommendation system. This model overcomes the problem of surplus information about movies and offers users only the relevant items by analyzing their interests and preferences. This work presents a hybrid movie recommendation system based on unsupervised clustering and supervised deep learning. Here, the dataset used is the IMDB movie reviews dataset. The clustering technique used is adaptive density-based clustering or adaptive DBSCAN. Next, the similarity calculation is performed using the cosine similarity metric. Finally, the Convolutional Neural Network (CNN) model integrated with the Adaptive Red Deer (ARD) Optimization offers the most relevant depictions on the movie list with higher accuracy. The implementation is performed on the MATLAB simulation environment. The system’s performance is measured in terms of accuracy, recall, precision, F-measure, RMSE and MAE.

Recommender systems are becoming a popular and important set of personalization techniques that assist individual users with navigating through the rapidly growing amount of information. A good recommender system should be able to not only find out the objects preferred by users, but also help users in discovering their personalized tastes. The former corresponds to high accuracy of the recommendation, while the latter to high diversity. A big challenge is to design an algorithm that provides both highly accurate and diverse recommendation. Traditional recommendation algorithms only take into account the contributions of similar users, thus, they tend to recommend popular items for users ignoring the diversity of recommendations. In this paper, we propose a recommendation algorithm by considering both the effects of similar and dissimilar users under the framework of collaborative filtering. Extensive analyses on three datasets, namely MovieLens, Netflix and Amazon, show that our method performs much better than the standard collaborative filtering algorithm for both accuracy and diversity.