Home OALib Journal OALib PrePrints Submit Ranking News My Lib FAQ About Us Follow Us+
 Title Keywords Abstract Author All
Search Results: 1 - 10 of 100 matches for " "
 Page 1 /100 Display every page 5 10 20 Item
 Computer Science , 2007, DOI: 10.1142/S0129183109013613 Abstract: In this paper, we propose a spreading activation approach for collaborative filtering (SA-CF). By using the opinion spreading process, the similarity between any users can be obtained. The algorithm has remarkably higher accuracy than the standard collaborative filtering (CF) using Pearson correlation. Furthermore, we introduce a free parameter $\beta$ to regulate the contributions of objects to user-user correlations. The numerical results indicate that decreasing the influence of popular objects can further improve the algorithmic accuracy and personality. We argue that a better algorithm should simultaneously require less computation and generate higher accuracy. Accordingly, we further propose an algorithm involving only the top-$N$ similar neighbors for each target user, which has both less computational complexity and higher algorithmic accuracy.
 Journal of Computer and Communications (JCC) , 2014, DOI: 10.4236/jcc.2014.214006 Abstract: Short text, based on the platform of web2.0, gained rapid development in a relatively short time. Recommendation systems analyzing user’s interest by short texts becomes more and more important. Collaborative filtering is one of the most promising recommendation technologies. However, the existing collaborative filtering methods don’t consider the drifting of user’s interest. This often leads to a big difference between the result of recommendation and user’s real demands. In this paper, according to the traditional collaborative filtering algorithm, a new personalized recommendation algorithm is proposed. It traced user’s interest by using Ebbinghaus Forgetting Curve. Some experiments have been done. The results demonstrated that the new algorithm could indeed make a contribution to getting rid of user’s overdue interests and discovering their real-time interests for more accurate recommendation.
 International Journal of Engineering Science and Technology , 2011, Abstract: Collaborative filtering (CF) systems have been proven to be very effective for personalized and accurate recommendations. These systems are based on the Recommendations of previous ratings byvarious users and products. Since the present database is very sparse, the missing values are considered first and based on that, a complete prediction dataset are made. In this paper, some standardcomputational techniques are applied within the framework of Content-boosted collaborative filtering with imputational rating data to evaluate and produce CF predictions. The Content-boosted collaborative filtering algorithm uses either naive Bayes or means imputation, depending on the sparsity of the original CF rating dataset. Results are presented and shown that this approach performs better than a traditional content-based predictor and collaborative filters.
 Computer Science , 2014, Abstract: The quality of an induced model by a learning algorithm is dependent on the quality of the training data and the hyper-parameters supplied to the learning algorithm. Prior work has shown that improving the quality of the training data (i.e., by removing low quality instances) or tuning the learning algorithm hyper-parameters can significantly improve the quality of an induced model. A comparison of the two methods is lacking though. In this paper, we estimate and compare the potential benefits of filtering and hyper-parameter optimization. Estimating the potential benefit gives an overly optimistic estimate but also empirically demonstrates an approximation of the maximum potential benefit of each method. We find that, while both significantly improve the induced model, improving the quality of the training set has a greater potential effect than hyper-parameter optimization.
 计算机应用研究 , 2011, Abstract: This paper presented a collaborative filtering algorithm based on a new similarity definition, namely meta similarity. By considering the vector correlation of the user similarity matrix, the meta similarity between any two users could be obtained. Even the relation of two users who had no common collected items could be investigated by it. Combining with the initial similarity with a tunable parameter, the integrated similarity could reflect relations between users more properly. Numerical results indicate that the algorithmic accuracy, measured by the average ranking score, and precision and recall is improved greatly.
 罗圣美, 林运祯, 叶小伟, 文海龙 Hans Journal of Data Mining (HJDM) , 2013, DOI: 10.12677/HJDM.2013.31003 Abstract: 协同过滤算法是现在个性化推荐领域流行的算法。对常见的推荐问题，协同过滤算法已有成熟的实现。单类协同过滤问题是推荐领域的一个新问题，其数据特征导致其不适用于常见的协同过滤算法。本文研究了基于加权矩阵分解的单类协同过滤算法，并对其进行基于迁移学习的改进。通过在真实数据集上的验证，证明其效果优于传统的单类协同过滤算法。 Collaborative filtering is a useful algorithm for problems of personalized recommendation. For these prob-lems, there are many mature collaborative filtering algorithms. One class collaborative filtering is a new field of per-sonalized recommendation. Because of its data characteristics, common collaborative filtering algorithms have a lot of defects in the field of one class collaborative filtering. We studied the algorithm based on weighted matrix decomposi-tion, and optimized this algorithm by transfer learning. We prove the improvement of this optimization by experiments.
 Journal of Multimedia , 2010, DOI: 10.4304/jmm.5.6.596-604 Abstract: Content-based image retrieval is a very dynamic study field, and in this field, how to improve retrieval speed and retrieval accuracy is a hot issue. The retrieval performance can be improved when applying relevance feedback to image retrieval and introducing the participation of people to the retrieval process. However, as for many existing image retrieval methods, there are disadvantages of relevance feedback with information not being fully saved and used, and their accuracy and flexibility are relatively poor. Based on this, the collaborative filtering technology was combined with relevance feedback in this study, and an improved relevance feedback algorithm based on collaborative filtering was proposed. In the method, the collaborative filtering technology was used not only to predict the semantic relevance between images in database and the retrieval samples, but to analyze feedback log files in image retrieval, which can make the historical data of relevance feedback be fully used by image retrieval system, and further to improve the efficiency of feedback. The improved algorithm presented has been tested on the content-based image retrieval database, and the performance of the algorithm has been analyzed and compared with the existing algorithms. The experimental results showed that, compared with the traditional feedback algorithms, this method can obviously improve the efficiency of relevance feedback, and effectively promote the recall and precision of image retrieval.
 控制理论与应用 , 2008, Abstract: In this paper, an adaptive estimation algorithm is proposed for non-linear dynamic systems with unknown static parameters based on combination of particle filtering and Simultaneous Perturbation Stochastic Approximation (SPSA) technique. The estimations of parameters are obtained by maximum-likelihood estimation and sampling within particle filtering framework, and the SPSA is used for stochastic optimization and to approximate the gradient of the cost function. The proposed algorithm achieves combined estimation of dynamic state and static parameters of nonlinear systems. Simulation result demonstrates the feasibility and efficiency of the proposed algorithm.
 计算机应用 , 2007, Abstract: Collaborative filtering is the most widely used recommendation technology in the personalized recommendation system. However, the user's interests in different time have been taken into equal consideration with the method being used, which leads to the lack of effectiveness in the given period of time. In view of this problem, this paper presented an improved collaborative filtering algorithm to make the click interests, approaching the gathering time, have bigger weight in the recommendation process, thereby to improve the accuracy of the recommendation.
 Songjie Gong Journal of Software , 2010, DOI: 10.4304/jsw.5.7.745-752 Abstract: Personalized recommendation systems can help people to find interesting things and they are widely used with the development of electronic commerce. Many recommendation systems employ the collaborative filtering technology, which has been proved to be one of the most successful techniques in recommender systems in recent years. With the gradual increase of customers and products in electronic commerce systems, the time consuming nearest neighbor collaborative filtering search of the target customer in the total customer space resulted in the failure of ensuring the real time requirement of recommender system. At the same time, it suffers from its poor quality when the number of the records in the user database increases. Sparsity of source data set is the major reason causing the poor quality. To solve the problems of scalability and sparsity in the collaborative filtering, this paper proposed a personalized recommendation approach joins the user clustering technology and item clustering technology. Users are clustered based on users’ ratings on items, and each users cluster has a cluster center. Based on the similarity between target user and cluster centers, the nearest neighbors of target user can be found and smooth the prediction where necessary. Then, the proposed approach utilizes the item clustering collaborative filtering to produce the recommendations. The recommendation joining user clustering and item clustering collaborative filtering is more scalable and more accurate than the traditional one.
 Page 1 /100 Display every page 5 10 20 Item