Collaborative Recommendation Method Research Articles

A collaborative filtering recommendation method based on emotional evaluation relations

A collaborative filtering recommendation method based on emotional evaluation relations

Similarity-based Graph Convolution Collaborative Recommendation Approach

With the rapid and iterative development of science and technology, a large amount of information exceeds the range that can be accepted, processed, or effectively utilized by an individual or a system, and recommendation algorithms can, to a certain extent, solve such problems, but traditional recommendation algorithms do not have a good solution to the problems related to data sparsity and recommendation accuracy. A similarity-based graph convolutional neural collaborative recommendation method (GCSCF) is proposed. The similarity algorithm based on the attributes of the item is used to find the item with the highest similarity and has not interacted with the current user, and this item is set to interact with the current user. The relevant interaction information of the user and the item is converted into relative feature vectors; the feature vectors are propagated using a graph convolutional neural network to aggregate the localized information, and the weight coefficients based on the item ratings are normalized to reduce the noise caused by the information aggregation. Comparative experiments are conducted on two public datasets, MovieLens-1M and Movielens-100K, with five baseline models on the set, and using Recall, Normalized Discounted Cumulative Gain (NDCG), and Precision as the evaluation metrics, and the results of the experiments show that the performance of the proposed social recommendation model better than other models.

Graph convolutional collaborative filtering recommendation method based on temporal information during node aggregation process

Graph Convolutional Networks (GCN) are prevalent techniques in collaborative filtering recommendations. However, current GCN-based approaches for collaborative filtering recommendation have limitations in effectively embedding neighboring nodes during node and neighbor information aggregation. Furthermore, weight allocation for the user (or item) representations after convolution of each layer is too uniform. To resolve these limitations, we propose a new Graph Convolutional Collaborative Filtering recommendation method based on temporal information during the node aggregation process (TA-GCCF). The method aggregates and propagates information using Gated Recurrent Units, while dynamically updating features based on the timing and sequence of interactions between nodes and their neighbors. Concurrently, we have developed a convolution attention coefficient to ascertain the significance of embedding at distinct layers. Experiments on three benchmark datasets show that our method significantly outperforms the comparison methods in the accuracy of prediction.

Application Analysis of English Personalized Learning Based on Large-scale Open Network Courses

In the context of Big data, large-scale open online courses increase learning paths for learners, but in the face of countless high-quality curriculum resources, it is easy for derivative learners to face the dilemma of rich curriculum resources but difficult to choose resources, which leads to information maze for learners. How to help learners quickly and accurately find their own learning resources in the explosive growth of MOOC resources is an urgent problem in the field of education Big data. However, the traditional Collaborative filtering recommendation technology does not perform well when dealing with sparse data and cold start. The recommendation content is repeated and can not effectively deal with high-dimensional and nonlinear data of online learning users, resulting in low efficiency of resource recommendation. Therefore, the study adopts a deep belief network (DBN) to construct a personalized resource recommendation model. The model combines the learner behavior characteristics with the curriculum resource content attribute characteristics to form the learner feature vector. The parameters of the model are adjusted according to the characteristics of learners. Through experiments, the proposed model has shown good performance. The experiment explored the effects of training set size, learner characteristics, and GPU on model performance. The experimental results show that when the training set proportion is 100%, the RMSE, Accuracy, Recall, and F1 values of the model are 0.76, 0.946, 0.957, and 0.951, respectively. When the model is trained using a training set containing learner features, the RMSE, Accuracy, Recall, and F1 values of the model are 0.75, 0.962, 0.908, and 0.958, respectively. After using GPU to accelerate the model, the running time of the model decreased from 360 minutes to 90 minutes. The results indicate that the model cannot effectively mine data information when the degree of correlation between sample information is low. The richer the relationships between samples, the better the performance of the model. Simultaneously learning hunger feature vectors and learner behavior feature vectors for training can significantly improve the recommendation accuracy of the model. The main contribution of this study is to propose a recommendation method based on DBN classification to replace traditional similarity calculation methods, using DBN's efficient feature abstraction and feature extraction capabilities to fully explore learners' interest and preference for course resources. In addition, in view of the common problems of cold start and data sparsity in traditional Collaborative filtering recommendation methods, the research deeply mines the characteristics of learners' Demographics and curriculum resources' content attributes, and constructs a learner interest model based on DBN combined with learners' behavior characteristics, which effectively solves the problems of cold start and data sparsity, as well as the inaccurate expression of learners' interest preferences for curriculum resources.

A Locality-Sensitive Hashing based Collaborative Recommendation Method for Responsible AI Driven Recommender Systems

A Locality-Sensitive Hashing based Collaborative Recommendation Method for Responsible AI Driven Recommender Systems

Construction of personalized learning service system based on deep learning and knowledge graph

Abstract With the development of information technology, it is difficult for people to find effective information from the massive Internet information, and personalized services are in short supply. This paper explores the essential association between knowledge graphs and deep learning and examines the construction process of knowledge graphs. Based on the content and collaborative filtering recommendation method, the theoretical method of personalized learning service recommendation is being studied. Knowledge graphs are introduced in collaborative filtering to accurately portray user and item vector representations. The knowledge graph is used to represent the TransR model, forming an entity space and a multivariate relationship space. The students’ demand for personalized learning service systems is analyzed from the user’s point of view, the personalized learning service system is constructed, and the accuracy of the personalized learning service system constructed in this paper is analyzed. The accuracy rate of DLKG is 0.1425 when K=10, which is the highest among the five models. In the effect of personalized teaching, the mean value of the metacognitive ability of the students in the experimental class is higher than that of the control class by 0.4155, 0.0261, 0.3995, 0.7967, respectively, and the experiment proves that the personalized learning service system based on the DLKG model proposed in this paper is more effective.

Research on the integration path of cultural creative industry and tourism industry based on collaborative filtering recommendation algorithm

Abstract In order to increase regional economic benefits and attract more tourists for local economic development. This paper adopts different ways to evolve industrial integration, enrich and expand the theoretical system of the industrial economy, establish new tourism strategies to meet personalized and diversified needs and make the tourism sector achieve industrial integration. Useful data are selected to filter cultural and creative, and tourism industry information by mining hidden information of coding sequences and finding support data items. Using collaborative filtering recommendation method to obtain the preference degree of fused data items, populate the data according to behavioral speculation, calculate the utility matrix similarity, and clarify the development of industry fusion path in the context of big data. The results show that the average error rate after fusion does not exceed 4%, and the fusion path of cultural and creative industries and the tourism industry improve the regional economy and the overall local living conditions.

Effective healthcare service recommendation with network representation learning: A recursive neural network approach

Recently, recommender systems have been combined with healthcare systems to recommend needed healthcare items for both patients and medical staff. By monitoring the patients’ states, healthcare services and their consumed smart medical objects can be recommended to a medical team according to the patient’s critical situation and requirements. However, a common drawback of the few existing solutions lies in the limited modeling of the healthcare information network. In addition, current solutions do not consider the typed nature of healthcare items. Moreover, existing healthcare recommender systems lack flexibility, and none of them offers re-configurable healthcare workflows to medical staff. In this paper, we take advantage of collaborative filtering and representation learning principles, by proposing a method for the recommendation of healthcare services. These latter follow a predefined execution pattern, i.e. treatment/medication workflow, that is determined by our framework depending on the patient’s state. To achieve this goal, we model the healthcare information network as a knowledge graph. This latter, based on an incremental learning method, is then transformed into a cuboid space to facilitate its processing. That is by learning latent representations of its content (e.g., smart objects, healthcare services, patients symptoms, etc.). Finally, a collaborative recommendation method is defined to select the high-quality healthcare services that will be composed and executed according to a determined workflow model. Experimental results have proven the efficiency of our solution in terms of recommended services’ quality.

Research on personalized recommendation method of preschool e-learning resources

Abstract E-learning is a very popular learning method at this stage, and Internet learners need to invest much time to retrieve the required e-learning network resources. In building the existing e-learning resources, the cognitive level, thinking ability, learning style, and other factors of learners are not considered. For this challenge, the paper gives the recommendation methods of the collaborative filtering algorithm model and learner model, compares the advantages and disadvantages of these models and traditional recommendation methods using the data of mean square error, response time, and accuracy, and examines the students’ suggestions for this recommendation method in the field. The collaborative filtering recommendation technique had the highest completion rate, above 0.7 for 5-20 recommendations and above 0.75 and 0.8 for 20-35. The rate of the learner model is stable between 0.7 and 0.8. In contrast, the detection rate of traditional methods is between 0.6 and 0.7. The collaborative filtering recommendation method outperforms the other two methods regarding completeness, response time, accuracy, and F1 value. In terms of satisfaction, the collaborative filtering recommendation algorithm had the highest satisfaction rate, followed by the learner model; the collaborative filtering method had the best results in improving learning performance. Therefore, personalized recommendation methods can improve learning efficiency and reduce retrieval time, and the collaborative filtering recommendation method is the most effective among the two methods proposed in this paper.

Collaborative variational factorisation machine for recommender system

At present, the recommendation system is confronting the huge challenge of data sparsity and high complexity of algorithm. Like the traditional collaborative filtering recommendation methods, they are difficult to adapt to the data sparse environment, resulting in low prediction accuracy. To address the issues, this paper presents a novel factorisation machine based on Collaborative filtering framework called collaborative variational factorisation machine (CVFM) that considers the user-user relation with the interaction data for Recommender systems. First, the user-item explicit ratings are used to build the user-user relationship by the similarity calculation. Next, we develop a variational factorisation machine to exploit the inherent relationship of latent variables from interaction information. The experimental results on three different sparsity datasets show that the presented CVFM is superior to other popular method in prediction accuracy, at the same time, maintain the stability of our algorithm with dealing with sparse data.

E-commerce collaborative filtering recommendation method based on social network user relationship

E-commerce collaborative filtering recommendation method based on social network user relationship

E-commerce Collaborative Filtering Recommendation Method Based on Social Network User Relationship

E-commerce Collaborative Filtering Recommendation Method Based on Social Network User Relationship

A Graph Convolution Collaborative Filtering Integrating Social Relations Recommendation Method

Traditional collaborative filtering recommendation algorithms only consider the interaction between users and items leading to low recommendation accuracy. Aiming to solve this problem, a graph convolution collaborative filtering recommendation method integrating social relations is proposed. Firstly, a social recommendation model based on graph convolution representation learning and general collaborative filtering (SRGCF) is constructed; then, based on this model, a social relationship recommendation algorithm (SRRA) is proposed; secondly, the algorithm learns the representations of users and items by linear propagation on the user–item bipartite graph; then the user representations are updated by learning the representations with social information through the neighbor aggregation operation in the social network to form the final user representations. Finally, the prediction scores are calculated, and the recommendation list is generated. The comparative experimental results on four real-world datasets show that: the proposed SRRA algorithm performs the best over existing baselines on Recall@10 and NDCG@10; specifically, SRRA improved by an average of 4.40% and 9.62% compared to DICER and GraphRec, respectively, which validates that the proposed SRGCF model and SRRA algorithm are reasonable and effective.

A privacy‐preserving recommendation method with clustering and locality‐sensitive hashing

AbstractNowadays, there is a significant increase in information, resulting in information overload. Recommendation systems have been widely adopted, and they can help users find information relevant to their interests. However, a malicious attacker can infer users' private information via recommendations. To solve problems of data sparseness, enormous high‐dimensional data, the cold start problem and privacy protection in an intelligent recommender system, this study proposes a privacy‐preserving collaborative filtering recommendation method with clustering and locality‐sensitive hashing. First, we cluster users according to their characteristic information to obtain sub‐rating matrices. We use the latent factor model to predict and fill in the missing ratings in those matrices. Second, we combine the sub‐rating matrices into a complete rating matrix, subsequently, we obtained the neighbors of the target user by analyzing the similarity of the users. We use a locality‐sensitive hashing algorithm to reduce the dimensionality of the user rating data and build an index that could quickly obtain the neighbors of the target user. Finally, we predict the target user's ratings and provide recommendations to the target user. Through experiments, our study shows that our method can deal with the problems of data sparseness and cold start problems well and the accuracy of the intelligent recommendation system has been improved. In addition, we use hash techniques to search for the neighbors, which effectively protects the privacy of the user.

Safety tag generation and training material recommendation for construction workers: a persona-based approach

PurposeEffective construction safety training has been considered to play a significant role in reducing the incidence of accidents. However, the current safety training methods pay less attention to the relationship between workers' personalized characteristics and their learning needs, which results in workers' low learning participation and poor training effect. The purpose of this paper is to improve the participation and effect of safety training for construction workers with a persona-based approach.Design/methodology/approachThis paper presents a persona-based approach to safety tag generation and training material recommendation. By extracting the demographic characteristics and behavior patterns tags of construction workers, a neural network algorithm is introduced to calculate the learning needs tags of workers, and the collaborative filtering recommendation method is integrated to enrich the innovation of recommendation results. Offline experiments and online experiments are designed to verify the rationality of the proposed method.FindingsThe results show that the learning needs of workers are closely related to their background. The proposed method can effectively improve workers' interest in materials and the training effect compared with conventional safety training methods. The research provides a theoretical and practical reference for promoting active safety management and achieving worker-centered safety management.Originality/valueFirst, a persona-based approach is introduced to establish a novel framework for solving the problem of personalized construction safety management. Second, an artificial intelligence algorithm is used to automatically extract the learning needs tag values and design a hybrid recommendation method for construction workers' personalized safety training. The collaborative filtering method is integrated to enrich the innovation of recommendation results.

Personalized Recommendation Algorithm of Tourist Attractions Based on Transfer Learning

With the development of information technology and the popularity of the Internet, the data on the network is growing exponentially. Information overload has become a significant issue for consumers seeking information. A recommendation system was created to detect users’ interests from huge amounts of data and to suit users’ specific information needs. Traditional collaborative filtering recommendation mostly uses scoring data for a recommendation, which has the problem of sparse data, which limits the performance of the recommendation system. On this basis, this paper studies the personalized recommendation algorithm of scenic spots with deep migration. Through the analysis of collaborative filtering recommendation methods, it is found that the traditional collaborative filtering methods only use scoring data for a recommendation, which has the problem of sparse data. Based on the vectorization of user interest, the similarity of user preference is calculated, and the matrix decomposition is carried out in cooperation with user implicit feedback, to integrate the knowledge transfer information into the matrix decomposition model, and make up for the lack of considering the attribute information of scenic spots in the matrix decomposition algorithm, and alleviate the problem of data sparsity. The findings of comparative trials suggest that the personalized scenic location recommendation approach proposed in this study, which is based on the depth migration algorithm, is effective. Compared with the benchmark recommendation method, the recommendation accuracy and recall rate has been improved to a certain extent.

Method of Ideological and Political Teaching Resources in Universities Based on School-Enterprise Cooperation Mode

The personalized recommendation system influences the recommendation of ideological and political teaching resources in universities, resulting in a high MAE score. As a result, under the school-enterprise collaboration paradigm, this study proposes a customised recommendation approach for ideological and political teaching resources in colleges and universities. The ideological and political teaching resource bank is developed against the backdrop of the teaching paradigm that combines universities and businesses. Learners’ browsing data history is gathered to create a learning interest model for them. A hybrid collaborative filtering recommendation method was devised, and a recommendation engine was established by Taste component, taking into account individualised resource recommendation needs and information entropy weight distribution mode. When compared to previous techniques, the developed customised recommendation method considerably enhances the recommendation quality of instructional resources and reduces MAE by 29% and 34%, respectively.

Collaborative Filtering Recommendation of Music MOOC Resources Based on Spark Architecture.

With the rapid development of MOOC platforms, MOOC resources have grown substantially, causing the problem of information overload. It is difficult for users to select the courses they need from a large number of MOOC resources. It is necessary to help users select the right music courses and at the same time make the outstanding music courses stand out. Recommendation systems are considered a more efficient way to solve the information overload problem. To improve the accuracy of the recommendation results of music MOOC resources, a mixed collaborative filtering recommendation algorithm based on Spark architecture is proposed. First, the user data and item data are modeled and scored by the collaborative filtering algorithm, then the tree structure of the XGBoost model and the features of regular learning are combined to predict the scores, and then the two algorithms are mixed to solve the optimal objective function to obtain the set of candidate recommendation data. Then, the frog-jumping algorithm is used to train the weighting factors, and the optimal combination of weighting factors is used as the training result of the samples to realize the data analysis of the mixed collaborative filtering recommendation algorithm. The experimental results in the music MOOC resource show that the average absolute error and root mean square error of the proposed method are 0.406 and 1.117, respectively, when the sparsity is 30%, which are lower than those of other existing collaborative filtering recommendation methods, with higher accuracy and execution efficiency.

Effective Hybrid Content-Based Collaborative Filtering Approach for Requirements Engineering

Requirements engineering (RE) is among the most valuable and critical processes in software development. The quality of this process significantly affects the success of a software project. An important step in RE is requirements elicitation, which involves collecting project-related requirements from different sources. Repositories of reusable requirements are typically important sources of an increasing number of reusable software requirements. However, the process of searching such repositories to collect valuable project-related requirements is time-consuming and difficult to perform accurately. Recommender systems have been widely recognized as an effective solution to such problem. Accordingly, this study proposes an effective hybrid content-based collaborative filtering recommendation approach. The proposed approach will support project stakeholders in mitigating the risk of missing requirements during requirements elicitation by identifying related requirements from software requirement repositories. The experimental results on the RALIC dataset demonstrate that the proposed approach considerably outperforms baseline collaborative filtering-based recommendation methods in terms of prediction accuracy and coverage in addition to mitigating the data sparsity and cold-start item problems.

A Hybrid Multi-Criteria Collaborative Filtering Model for Effective Personalized Recommendations

Recommender systems act as decision support systems in supporting users in selecting the right choice of items or services from a high number of choices in an overloaded search space. However, such systems have difficulty dealing with sparse rating data. One way to deal with this issue is to incorporate additional explicit information, also known as side information, to the rating information. However, this side information requires some explicit action from the users and often not always available. Accordingly, this study presents a hybrid multi-criteria collaborative filtering model. The proposed model exploits the multi-criteria ratings, implicit similarity, similarity transitivity and global reputation concepts to expand the space of potential recommenders. This expansion will enhance the prediction accuracy and coverage of the proposed model when applied to sparse data situations. To show effectiveness of the proposed model, a set of experiments are conducted on two real-world multi-criteria datasets, Yahoo! Movies and TripAdvisor. The experimental results demonstrate the superiority of the proposed model compared to a number of existing collaborative filtering-based recommendation methods under a variety of evaluation metrics.