While contemporary community-based recommendation algorithms based on a single community structure are more capable of processing large datasets than ever, they lack recommendation precision. This article proposes a c...
详细信息
While contemporary community-based recommendation algorithms based on a single community structure are more capable of processing large datasets than ever, they lack recommendation precision. This article proposes a collaborative filtering recommendation algorithm that integrates community structure and user implicit trust. The algorithm first applies a method based on the Gaussian function to fill the matrix of item ratings of users to alleviate data sparsity. It then uses the trust matrix to obtain the asymmetric trust relationship of the trustor and trustee, based on which the degree of users' implicit trust is calculated. The users are divided into communities based on the implicit trust degree to determine the influence among users more accurately. The algorithm then predicts the target user's rating using the ratings of users in the community to generate recommendations. To verify the performance of the proposed algorithm, we compared the proposed algorithm with three contemporary algorithms under the same conditions using FilmTrust datasets. The recommendation accuracy as well as the mean absolute error and root mean square error values of the proposed algorithm were better than those of the other four algorithms by approximately 14% and 4%, respectively. The experimental results demonstrate that the proposed algorithm can achieve better recommendation efficiency than existing algorithms.
Social network service (SNS) based recommendation system is playing an important role in searching information in the vast number of SNS community websites. In this paper, we propose a recommendation algorithm using i...
详细信息
ISBN:
(纸本)9781479906048;9781479906024
Social network service (SNS) based recommendation system is playing an important role in searching information in the vast number of SNS community websites. In this paper, we propose a recommendation algorithm using intimacy information based on the SNS after extracting information on the social network having relationship with the user. It is an algorithm to recommend the contents with high relationship based on the list of friends who are predicted as actually having close relationship with the user according to the familiarity after extracting the list of friends within the SNS using SNS information of the user.
With the development of mobile Internet, more and more users get information through mobile terminals. However, the current personalized recommendation system is lack of the ability to perceive the user's situatio...
详细信息
ISBN:
(纸本)9783030945541;9783030945534
With the development of mobile Internet, more and more users get information through mobile terminals. However, the current personalized recommendation system is lack of the ability to perceive the user's situation and provide personalized information recommendation service for users. For this reason, this paper proposes a context aware recommendation algorithm based on RFID application, which collects user resource category preference learning features by combining with context features, and integrates different category preferences for collaborative filtering personalized information recommendation. The RFID method is used to learn the user's preference for each resource category in different contexts, and then the category preference is combined with the traditional RFID context-aware recommendation algorithm to generate personalized information recommendations that are in line with the user's current context. Experiments show that the context-aware recommendation algorithm proposed in this paper based on RFID applications can improve the accuracy of recommendation.
In order to solve the sparsity of User-Item scoring matrix in collaborative filtering recommendation, proposed an algorithm that combining user's interests with item's quality to calculate ungraded items in ma...
详细信息
ISBN:
(纸本)9781467371896
In order to solve the sparsity of User-Item scoring matrix in collaborative filtering recommendation, proposed an algorithm that combining user's interests with item's quality to calculate ungraded items in matrix. By setting the weight of user and item, synthesized the value of missing value, which were used to replace the ungraded value in scoring matrix to calculate similarity. Experiment shows that the algorithm can improve the recommendation effect, and when the user's weight values 0.4, MAE reaches minimum, and recommendation quality reaches maximum.
With the huge amount of information available on the Internet, recommendation systems gained popularity over the years. Traditional recommendation algorithm usually uses collaborative filtering to determine user and i...
详细信息
ISBN:
(纸本)9781728113227
With the huge amount of information available on the Internet, recommendation systems gained popularity over the years. Traditional recommendation algorithm usually uses collaborative filtering to determine user and item similarity. However, data sparsity and overfitting affects the accuracy of the recommendation systems that lead to poor recommendation quality. This paper presents an enhanced recommendation algorithm based on modified user-based collaborative filtering to overcome the problem and improve the recommendation quality. The enhanced algorithm was compared to the traditional algorithm using the MovieLens dataset and evaluates its accuracy and performance using the Root Mean Square Error (RSME), Precision and Recall. The experimental results show that the enhanced algorithm outperforms the traditional algorithm and improves the accuracy of the recommendation.
In this paper the collaborative filtering recommendation and association rules are introduced firstly. Aiming at the shortcomings of basic collaborative filtering recommendation, this paper proposes an improved collab...
详细信息
ISBN:
(纸本)9783037858646
In this paper the collaborative filtering recommendation and association rules are introduced firstly. Aiming at the shortcomings of basic collaborative filtering recommendation, this paper proposes an improved collaborative filtering recommendation algorithm based on weighted association rules (CFRA-WAR). Finally the simulation experiments are carried out to verify the validity of the improved recommendation algorithm. The results of simulation experiments show that the recommendation accuracy of CFRA-WAR is superior to basic collaborative filtering recommendation algorithm, although the algorithm time of CFRA-WAR is a little longer.
With the rapid development of the information society, both the producers of teaching resources and students are facing enormous challenges. In this context, a recommendation system that can connect the producers and ...
详细信息
ISBN:
(纸本)9781450362658
With the rapid development of the information society, both the producers of teaching resources and students are facing enormous challenges. In this context, a recommendation system that can connect the producers and students of teaching resources to achieve a win-win situation is proposed. Collaborative filtering is currently the most successful recommendation technique. In this article the authors firstly introduced the collaborative filtering recommendation algorithm from the angle of user-based, project-based and model-based algorithms. Then they studied the current popular SVD and RBM recommendation algorithms mainly by comparing the experimental results of the basic project-based, SVD and RBM algorithms on the public data set movielens, and pointed out that the recommendation system will be a real-time, multi-modal trend in the age of educational informatization. The rapid development of education informatization has made MOOC, micro-classes and other teaching resources mainly relying on video resources to grow rapidly. The collaborative filtering recommendation algorithm expects to associate the most suitable high-quality resources with the most needed students.
Because the degree of mastering knowledge points in courses in traditional cognitive diagnostic models cannot be probabilistic, there are only two situations: mastery and non-mastery. Therefore, for the current resear...
详细信息
ISBN:
(数字)9781728199283
ISBN:
(纸本)9781728199283
Because the degree of mastering knowledge points in courses in traditional cognitive diagnostic models cannot be probabilistic, there are only two situations: mastery and non-mastery. Therefore, for the current research, the recommendations of knowledge points recommended by learners' learning behavior attributes are not fully considered to be insufficient, this paper proposes a curriculum knowledge point recommendation algorithm model based on learning diagnosis, the model comprehensively considers the learner's learning emotions, learner problem test conditions and knowledge point characteristics, and the film and television synthesis in the Chaoxing online teaching service platform the course learning data is tested to verify the effectiveness of the recommendation algorithm. The experimental results show that the effectiveness and accuracy of the recommendation algorithm model proposed in this paper can meet the learning needs of learners.
With the rapid development of the E-commerce, recommendation system and algorithm has become the relevant research hotspot. In this paper, we proposed a new recommendation algorithm,based on Multi-level Association Ru...
详细信息
ISBN:
(纸本)9783642286544
With the rapid development of the E-commerce, recommendation system and algorithm has become the relevant research hotspot. In this paper, we proposed a new recommendation algorithm,based on Multi-level Association Rules (MAR). The algorithm improves the precision and individuation degree of the recommendation,and significantly reduces the time needed for the recommendation. It mines the rule of the customer's choice of commodity,by using multi-level association rule, builds a model for choice prediction.
A real society has multiple social relationships between users, but existing social network recommendation algorithms often only introduce a social relationship into the recommendation system. This paper introduces a ...
详细信息
ISBN:
(纸本)9781728187129
A real society has multiple social relationships between users, but existing social network recommendation algorithms often only introduce a social relationship into the recommendation system. This paper introduces a variety of social relationships into the recommendation system based on a multi-subnet composite complex network model. Based on the analysis of the experimental results on the Epinions dataset, a recommendation algorithm introducing multiple social relationships has a significantly higher recommendation accuracy than a recommendation algorithm.
暂无评论