Collaborative Filtering Schemes Research Articles

UsCoTc: Improved Collaborative Filtering (CFL) recommendation methodology using user confidence, time context with impact factors for performance enhancement.

In today's society, time is considered more valuable than money, and researchers often have limited time to find relevant papers for their research. Identifying and accessing essential information can be a challenge in this situation. To address this, the personalized suggestion system has been developed, which uses a user's behavior data to suggest relevant items. The collaborative filtering strategy has been used to provide a user with the top research articles based on their queries and similarities with other users' questions, thus saving time by avoiding time-consuming searches. However, when rating data is abundant but sparse, the usual method of determining user similarity is relatively straightforward. Furthermore, it fails to account for changes in users' interests over time resulting in poor performance. This research proposes a new similarity measure approach that takes both user confidence and time context into account to increase user similarity computation. The experimental results show that the proposed technique works well with sparse data, and improves accuracy by 16.2% compared to existing models, especially during prediction. Furthermore, it enhances the quality of recommendations.

Movie Recommendation System using Machine Learning

Movies are very significant in our lives. It is one of the many forms of entertainment that we encounter in our daily lives. It is up to the individual to decide whatever type of film they choose to see, whether it is a comedy, romantic film, action film, or adventure film. However, the issue is locating acceptable content, as there is a large amount of information created each year. As a result, finding our favourite film is really difficult. The goal of this research is to improve the regular filtering technique's performance and accuracy. A recommendation system can be implemented using a variety of approaches. Content-based filtering and collaborative filtering strategies are employed in this work. The content-based filtering approach analyses the user's history/past behaviour and recommends a list of comparable movies depending on their input. K-NN algorithms and collaborative filtering are also employed in this paper to improve the accuracy of the results. Cosine similarity is utilised in this work to quickly discover comparable information. The correctness of the cosine angle is measured by cosine similarity. People may quickly find their favourite movie content thanks to all of this.

Improving Ant Collaborative Filtering on Sparsity via Dimension Reduction

Recommender systems should be able to handle highly sparse training data that continues to change over time. Among the many solutions, Ant Colony Optimization, as a kind of optimization algorithm modeled on the actions of an ant colony, enjoys the favorable characteristic of being optimal, which has not been easily achieved by other kinds of algorithms. A recent work adopting genetic optimization proposes a collaborative filtering scheme: Ant Collaborative Filtering (ACF), which models the pheromone of ants for a recommender system in two ways: (1) use the pheromone exchange to model the ratings given by users with respect to items; (2) use the evaporation of existing pheromone to model the evolution of users’ preference change over time. This mechanism helps to identify the users and the items most related, even in the case of sparsity, and can capture the drift of user preferences over time. However, it reveals that many users share the same preference over items, which means it is not necessary to initialize each user with a unique type of pheromone, as was done with the ACF. Regarding the sparsity problem, this work takes one step further to improve the Ant Collaborative Filtering’s performance by adding a clustering step in the initialization phase to reduce the dimension of the rate matrix, which leads to the results that K<<#users, where K is the number of clusters, which stands for the maximum number of types of pheromone carried by all users. We call this revised version the Improved Ant Collaborative Filtering (IACF). Experiments are conducted on larger datasets, compared with the previous work, based on three typical recommender systems: (1) movie recommendations, (2) music recommendations, and (3) book recommendations. For movie recommendation, a larger dataset, MoviesLens 10M, was used, instead of MoviesLens 1M. For book recommendation and music recommendation, we used a new dataset that has a much larger size of samples from Douban and NetEase. The results illustrate that our IACF algorithm can better deal with practical recommendation scenarios that handle sparse dataset.

Personalized real-time movie recommendation system: Practical prototype and evaluation

With the eruption of big data, practical recommendation schemes are now very important in various fields, including e-commerce, social networks, and a number of web-based services. Nowadays, there exist many personalized movie recommendation schemes utilizing publicly available movie datasets (e.g., MovieLens and Netflix), and returning improved performance metrics (e.g., Root-Mean-Square Error (RMSE)). However, two fundamental issues faced by movie recommendation systems are still neglected: first, scalability, and second, practical usage feedback and verification based on real implementation. In particular, Collaborative Filtering (CF) is one of the major prevailing techniques for implementing recommendation systems. However, traditional CF schemes suffer from a time complexity problem, which makes them bad candidates for real-world recommendation systems. In this paper, we address these two issues. Firstly, a simple but high-efficient recommendation algorithm is proposed, which exploits users' profile attributes to partition them into several clusters. For each cluster, a virtual opinion leader is conceived to represent the whole cluster, such that the dimension of the original user-item matrix can be significantly reduced, then a Weighted Slope One-VU method is designed and applied to the virtual opinion leader-item matrix to obtain the recommendation results. Compared to traditional clustering-based CF recommendation schemes, our method can significantly reduce the time complexity, while achieving comparable recommendation performance. Furthermore, we have constructed a real personalized web-based movie recommendation system, MovieWatch, opened it to the public, collected user feedback on recommendations, and evaluated the feasibility and accuracy of our system based on this real-world data.

Collaborative Filtering-Based Electricity Plan Recommender System

Owning to electricity market deregulation, residential customers now enjoy the freedom to choose their preferred electricity retailers. This paper investigates the application of recommender system, a fast-developing technique in machine learning, into the task of recommending electricity plans for the individual residential customer. Based on a collaborative filtering strategy, an electricity plan recommender system (EPRS) is developed. By providing easily obtainable data of some household appliances, residential customers of the EPRS are recommended with predicted ratings of different plans, which can provide effective guidance to customers in the selection of suitable plans and proper tariffs. Different numerical tests are carried out to evaluate the performance of the EPRS. The EPRS outperforms other strategies in the accuracy of recommendation result and is verified to be a promising solution to electricity plan recommendation task.

Clustering-Based Collaborative Filtering Using an Incentivized/Penalized User Model

Giving or recommending appropriate content based on the quality of experience is the most important and challenging issue in recommender systems. As collaborative filtering (CF) is one of the most prominent and popular techniques used for recommender systems, we propose a new clusteringbased CF (CBCF) method using an incentivized/penalized user (IPU) model only with the ratings given by users, which is thus easy to implement. We aim to design such a simple clustering-based approach with no further prior information while improving the recommendation accuracy. To be precise, the purpose of CBCF with the IPU model is to improve recommendation performance such as precision, recall, and F1 score by carefully exploiting different preferences among users. Specifically, we formulate a constrained optimization problem in which we aim to maximize the recall (or equivalently F1 score) for a given precision. To this end, users are divided into several clusters based on the actual rating data and Pearson correlation coefficient. Afterward, we give each item an incentive/penalty according to the preference tendency by users within the same cluster. Our experimental results show a significant performance improvement over the baseline CF scheme without clustering in terms of recall or F1 score for a given precision.

BRTCO: A Novel Boundary Recognition and Tracking Algorithm for Continuous Objects in Wireless Sensor Networks

Continuous objects monitoring and tracking, which aims to detect the invasion of unauthorized materials of interest, is one of the most prominent applications in wireless sensor networks. In this paper, we propose a novel boundary recognition and tracking algorithm for continuous objects (BRTCO) to ensure the efficiency of objects contour extraction. On precondition of assuring the tracking accuracy, a collaborative filtering scheme is proposed to minimize the number of boundary nodes. Also, in the phase of data transmission, we take the advantage of clustering to ensure energy efficiency. A report node selection mechanism is designed based on the competition of cluster heads. Simulation results demonstrate that BRTCO can significantly reduce the total energy consumption, the number of boundary nodes, and the number of report nodes.

Improving memory-based user collaborative filtering with evolutionary multi-objective optimization

The primary task of a memory-based collaborative filtering (CF) recommendation system is to select a group of nearest (similar) user neighbors for an active user. Traditional memory-based CF schemes tend to only focus on improving as much as possible the accuracy by recommending familiar items (i.e., popular items over the group). Yet, this may reduce the number of items that could be recommended and consequently weakens the chances of recommending novel items. To address this problem, it is desirable to consider recommendation coverage when selecting the appropriate group. This could help in simultaneously making both accurate and diverse recommendations. In this paper, we propose to focus mainly on the growing of the large search space of users’ profiles and to use an evolutionary multi-objective optimization-based recommendation system to pull up a group of profiles that maximizes both similarity with the active user and diversity between its members. In such manner, the recommendation system will provide high performances in terms of both accuracy and diversity. The experimental results on the Movielens benchmark and on a real-world insurance dataset show the efficiency of our approach in terms of accuracy and diversity compared to state-of-the-art competitors.

Three-Dimensional Segmentation of the Ex-Vivo Anterior Lamina Cribrosa From Second-Harmonic Imaging Microscopy.

The lamina cribrosa (LC) is a connective tissue in the posterior eye with a complex mesh-like trabecular microstructure, through which all the retinal ganglion cell axons and central retinal vessels pass. Recent studies have demonstrated that changes in the structure of the LC correlate with glaucomatous damage. Thus, accurate segmentation and reconstruction of the LC is of utmost importance. This paper presents a new automated method for segmenting the microstructure of the anterior LC in the images obtained via multiphoton microscopy using a combination of ideas. In order to reduce noise, we first smooth the input image using a 4-D collaborative filtering scheme. Next, we enhance the beam-like trabecular microstructure of the LC using wavelet multiresolution analysis. The enhanced LC microstructure is then automatically extracted using a combination of histogram thresholding and graph-cut binarization. Finally, we use morphological area opening as a postprocessing step to remove the small and unconnected 3-D regions in the binarized images. The performance of the proposed method is evaluated using mutual overlap accuracy, Tanimoto index, F-score, and Rand index. Quantitative and qualitative results show that the proposed algorithm provides improved segmentation accuracy and computational efficiency compared to the other recent algorithms.

A hybrid framework for social tag recommendation using context driven social information

Tagging images uploaded on the web is important as tags serve as the primary entities for future retrieval of these images. The numbers of images uploaded to the web via social networking sites is increasing in the era of smart phone technology and the internet. These images need to be tagged correctly which would ease its future retrieval. With the emergence of the Web 3.0 which is a standard for semantic web, a semantic tag recommender is desirable. In this paper, a context aware social tagger is proposed which recommends high quality tags by using varied contexts of social information. A semantic collaborative filtering strategy is proposed to make the social tagger semantics compliant. The social tagger also encompasses an intelligent agent, driven by second order co-occurrence pointwise mutual information strategy to increase the relevance and quality of the recommended tags. The proposed social tagger yields an average accuracy of 84.04%.

Merging user and item based collaborative filtering to alleviate data sparsity

Memory based algorithms, generally referred as similarity based Collaborative Filtering (CF) algorithm, is one of the most widely accepted approaches to provide service recommendations. It provides personalized and automated suggestions to customers to select variety of products. Memory based algorithms mainly have two kinds of algorithms: User-based and Item-based algorithms. The User-based CF algorithm recommends items by finding similar users. Contrary to User-based CF, an Item-based CF algorithm recommends items by finding similar items. The core of memory based CF technologies is to calculate similarity among users or items. However, due to inherent sparsity, a large number of entries (ratings) in user-item rating matrix are missing. This results in only few available ratings to make prediction for the unknown ratings. This results in poor prediction quality of the CF algorithm. In this paper a hybrid approach is presented that combines user-based CF and item-based CF. It also leverage the biclustering technique to reduce the dimensionality. The biclustering helps to cluster all users/items into several groups. These clusters are then used to measure users/items similarities based on their respective parent groups. To obtain individual prediction, it adopts the user-based and item-based CF schemes based on the computed similarity respectively. Finally it combines the resultant predictions of each model to make final predictions. Interestingly, experiments demonstrated that the proposed approach outperforms the traditional user-based, item-based and some state of the art recommendation approaches in terms of accuracy of prediction and quality of recommendations.

Segment-Focused Shilling Attacks against Recommendation Algorithms in Binary Ratings-based Recommender Systems

As one of the most successful approaches to building recommender systems, collaborative filtering (CF) uses the known preferences of a group of users to make recommendations or predictions of the unknown preferences for other users. However, the open nature of collaborative filtering recommender systems allows attackers who inject biased profile data to have a significant impact on the recommendations produced. Shilling attacks against numeric ratings-based CF schemes have been extensively studied. To the best of our knowledge, there is rare study about how to attack binary ratings-based recommendation systems. Hence, shilling attacks strategies against binary ratings-based recommendation algorithms need to be further investigated. The empirical results obtained from Movielens dataset show that the segment attack, which is easy to mount, affects strongly against BU-CF algorithm and BI-CF algorithm, and the attack size and the filler size don’t have the same sensitivity in attack effect.

Randomization-based Privacy-preserving Frameworks for Collaborative Filtering

Randomization-based privacy protection methods are widely used in collaborative filtering systems to achieve individual privacy. The basic idea behind randomization utilized in collaborative filtering schemes is to add randomness to original data in such a way so that required levels of accuracy and privacy can be achieved. Although there are various studies on privacy-preserving collaborative filtering using randomization, there are no well-defined privacy-preserving frameworks for collaborative filtering algorithms based on randomization. In this paper, we present eight randomization-based privacy-preserving frameworks for privacy protection in collaborative filtering schemes. We first group privacy-preserving methods into two broad categories. We then classify them based on private data. Final grouping is done while considering varying privacy concerns of individual users. The frameworks can be chosen according to individual users’ expectations and be utilized for privacy protection. The well-defined privacy-preserving frameworks form a basis for privacy protection based on randomized perturbation and randomized response techniques in collaborative filtering studies.

Privacy-Preserving Two-Party Collaborative Filtering on Overlapped Ratings

To promote recommendation services through prediction quality, some privacy-preserving collaborative filtering solutions are proposed to make e-commerce parties collaborate on partitioned data. It is almost probable that two parties hold ratings for the same users and items simultaneously; however, existing two-party privacy-preserving collaborative filtering solutions do not cover such overlaps. Since rating values and rated items are confidential, overlapping ratings make privacy-preservation more challenging. This study examines how to estimate predictions privately based on partitioned data with overlapped entries between two e-commerce companies. We consider both user-based and item-based collaborative filtering approaches and propose novel privacy-preserving collaborative filtering schemes in this sense. We also evaluate our schemes using real movie dataset, and the empirical outcomes show that the parties can promote collaborative services using our schemes.

Cold-Start Recommendation Using Bi-Clustering and Fusion for Large-Scale Social Recommender Systems

Social recommender systems leverage collaborative filtering (CF) to serve users with content that is of potential interesting to active users. A wide spectrum of CF schemes has been proposed. However, most of them cannot deal with the cold-start problem that denotes a situation that social media sites fail to draw recommendation for new items, users or both. In addition, they regard that all ratings equally contribute to the social media recommendation. This supposition is against the fact that low-level ratings contribute little to suggesting items that are likely to be of interest of users. To this end, we propose bi-clustering and fusion (BiFu)-a newly-fashioned scheme for the cold-start problem based on the BiFu techniques under a cloud computing setting. To identify the rating sources for recommendation, it introduces the concepts of popular items and frequent raters. To reduce the dimensionality of the rating matrix, BiFu leverages the bi-clustering technique. To overcome the data sparsity and rating diversity, it employs the smoothing and fusion technique. Finally, BiFu recommends social media contents from both item and user clusters. Experimental results show that BiFu significantly alleviates the cold-start problem in terms of accuracy and scalability.

Top-[formula omitted] followee recommendation over microblogging systems by exploiting diverse information sources

Followee recommendation plays an important role in information sharing over microblogging platforms. We frame this problem as a top-k ranking in collaborative filtering (CF). The difficulty is that explicit user-to-user ratings are not available on microblogging systems. Thus existing CF schemes are not applicable to followee recommendation over microblogging systems. To solve this problem, in this paper, we propose a novel followee ranking scheme using a variation of the latent factor model, which leverages implicit users’ feedback including both tweet content and social relation information. To achieve good top-k recommendation, we introduce a rank-based criterion to latent factor model (LFM). The main obstacle for training the model parameters is the non-smoothness of the objective function of LFM, which makes traditional parameter optimization methods infeasible. To tackle with the problem, we further design a smooth version of the objective function. We conduct comprehensive experiments on a large-scale dataset collected from Sina Weibo, the most popular microblogging system in China and a real world experiment on the Amazon Mechanical Turk CrowdSourcing platform to evaluate the performance of our design. The results show that our scheme greatly outperforms existing schemes in terms of precision and top-k ranking by 46.8% and 32.8%, respectively.

A Particle Swarm Optimization Approach to Fuzzy Case-based Reasoning in the Framework of Collaborative Filtering

The particle Swarm Optimization (PSO) algorithm, as one of the most effective search algorithm inspired from nature, is successfully applied in a variety of fields and is demonstrating fairly immense potential for development. Recently, researchers are investigating the use of PSO algorithm in the realm of personalized recommendation systems for providing tailored suggestions to users. Collaborative filtering (CF) is the most promising technique in recommender systems, providing personalized recommendations to users based on their previously expressed preferences and those of other similar users. However, data sparsity and prediction accuracy are the major concerns related to CF techniques. In order to handle these problems, this paper proposes a novel approach to CF technique by employing fuzzy case-based reasoning (FCBR) augmented with PSO algorithm, called PSO/FCBR/CF technique. In this method, the PSO algorithm is utilized to estimate the features importance and assign their weights accordingly in the process of fuzzy case-based reasoning (FCBR) for the computation of similarity between users and items. In this way, PSO embedded FCBR algorithm is applied for the prediction of missing values in user-item rating matrix and then CF technique is employed to generate recommendations for an active user. The experimental results clearly reveal that the proposed scheme, PSO/FCBR/CF, deals with the problem of sparsity as well as improves the prediction accuracy when compared with other state of the art CF schemes.

Enhanced New User Recommendations based on Quantitative Association Rule Mining

In the era of information explosion, how to provide tailored suggestions to a new user is a major concern for collaborative filtering (CF) based recommender systems. The CF recommender system performs very poorly for a new user with very poor profile information. Therefore, we investigate the use of quantitative association rules (QARs) for making recommendations to a new user by exploiting the cold user data which is readily available such as age, gender, occupation, etc. and ratings of items available in the historical data set. The proposed recommendation method, called QARF (QAR based filtering scheme), extracts relationships between readily available information of users and items, and the rating values. Additionally, QARs are extracted during offline processing which optimizes the online computation cost. The discovered rules are then employed during online processing in order to generate recommendations for a new user. Moreover, the QARF recommendation scheme is combined with CF, namely QARF/CF, to further improve recommendation accuracy. Proposed approaches QARF and QARF/CF are evaluated on the platform of MovieLens dataset. Experimental results demonstrate that the proposed schemes enhance new user recommendations and outperform other state of the art CF schemes.

An improved privacy-preserving DWT-based collaborative filtering scheme

Collaborative filtering (CF) is one of the most efficient techniques to produce personalized recommendations and to deal with the information overload of modern times. Although CF techniques have immensely useful filtering capabilities, many CF systems have challenging problems like scalability, accuracy, and privacy. One approach to enhance scalability of such systems is to apply discrete wavelet transformation (DWT) techniques. DWT-based CF schemes significantly overcome the scalability problem. However, they fail to protect individual users’ privacy. Moreover, although such schemes provide accurate predictions, the quality of the recommendations provided by DWT-based CF schemes can be further improved by applying some preprocessing methods.In this study, we propose privacy-preserving schemes to produce accurate predictions based on DWT efficiently without deeply exposing customers’ privacy. We also recommend methods to order items before applying DWT to boost accuracy. After evaluating our schemes in terms of privacy and supplementary costs, we perform real data-based experiments to scrutinize the proposed schemes in terms of accuracy. Experimental results show that our privacy-preserving methods are able to offer recommendations with decent accuracy. Moreover, our outcomes show that our methods utilized to sort items improve accuracy. We finally provide some suggestions and explain future works.

A Collaborative Filtering Approach for Protein-Protein Docking Scoring Functions

A protein-protein docking procedure traditionally consists in two successive tasks: a search algorithm generates a large number of candidate conformations mimicking the complex existing in vivo between two proteins, and a scoring function is used to rank them in order to extract a native-like one. We have already shown that using Voronoi constructions and a well chosen set of parameters, an accurate scoring function could be designed and optimized. However to be able to perform large-scale in silico exploration of the interactome, a near-native solution has to be found in the ten best-ranked solutions. This cannot yet be guaranteed by any of the existing scoring functions.In this work, we introduce a new procedure for conformation ranking. We previously developed a set of scoring functions where learning was performed using a genetic algorithm. These functions were used to assign a rank to each possible conformation. We now have a refined rank using different classifiers (decision trees, rules and support vector machines) in a collaborative filtering scheme. The scoring function newly obtained is evaluated using 10 fold cross-validation, and compared to the functions obtained using either genetic algorithms or collaborative filtering taken separately.This new approach was successfully applied to the CAPRI scoring ensembles. We show that for 10 targets out of 12, we are able to find a near-native conformation in the 10 best ranked solutions. Moreover, for 6 of them, the near-native conformation selected is of high accuracy. Finally, we show that this function dramatically enriches the 100 best-ranking conformations in near-native structures.