MONERS: A news recommender for the mobile web

Mobile web news services, which served by mobile service operators collecting news articles from diverse news contents providers, provide articles sorted by category or on the basis of attributes, such as the time at which they were posted. The mobile web should provide easy access to the categories or news contents preferred by users because user interface of wireless devices, particularly cell phones is limited for browsing between contents.This paper presents a mobile web news recommendation system (MONERS) that incorporates news article attributes and user preferences with regard to categories and news articles. User preference of news articles are estimated by aggregating news article importance and recency, user preference change, and user segment’s preference on news categories and articles. Performance of MONERS was tested in an actual mobile web environment; news organized by category had more page hits, while recommended news had a higher overall article read ratio.     

Group affinity based social trust model for an intelligent movie recommender system

As many researchers have taken an interest in social networks with the development of the user-generated web, trust management and its application have come into the spotlight. User information that is extracted by behavior patterns and user profiles provides the essential relationship between individuals. In this paper, we propose an intelligent movie recommender system with a social trust model. The proposed system is based on a social network for analyzing social relationships between users and generated group affinity values with user profiles. In experiments, the performance of this system is evaluated with precision-recall and F-measures.     

Social movie recommender system based on deep autoencoder network using Twitter data

Neural Computing and Applications - Recommender systems attempt to provide effective suggestions to each user based on their interests and behaviors. These recommendations usually match the...     

Probabilistic approach for QoS-aware recommender system for trustworthy web service selection

We present a QoS-aware recommender approach based on probabilistic models to assist the selection of web services in open, distributed, and service-oriented environments. This approach allows consumers to maintain a trust model for each service provider they interact with, leading to the prediction of the most trustworthy service a consumer can interact with among a plethora of similar services. In this paper, we associate the trust in a service to its performance denoted by QoS ratings instigated by the amalgamation of various QoS metrics. Since the quality of a service is contingent, which renders its trustworthiness uncertain, we adopt a probabilistic approach for the prediction of the quality of a service based on the evaluation of past experiences (ratings) of each of its consumers. We represent the QoS ratings of services using different statistical distributions, namely multinomial Dirichlet, multinomial generalized Dirichlet, and multinomial Beta-Liouville. We leverage various machine learning techniques to compute the probabilities of each web service to belong to different quality classes. For instance, we use the Bayesian inference method to estimate the parameters of the aforementioned distributions, which presents a multidimensional probabilistic embodiment of the quality of the corresponding web services. We also employ a Bayesian network classifier with a Beta-Liouville prior to enable the classification of the QoS of composite services given the QoS of its constituents. We extend our approach to function in an online setting using the Voting EM algorithm that enables the estimation of the probabilities of the QoS after each interaction with a web service. Our experimental results demonstrate the effectiveness of the proposed approaches in modeling, classifying and incrementally learning the QoS ratings.     

Algorithms for integrating PostgreSQL with the semantic web

This paper addresses the integration of the PostgreSQL database management system (DBMS) with the Semantic Web. Integration algorithms based on the use of the DBMS capabilities that do not introduce changes in SQL are proposed. An integration based on one of the algorithms supporting main formats of the Semantic Web is presented. The proposed algorithms can be implemented in different DBMSs supporting triggers (or rules), table functions, and indexing.     

A topic-based recommender system for electronic marketplace platforms

A large number of items are described, and subsequently bought and sold every day in auction marketplaces across the web. The amount of information and the number of available items makes finding what to buy as well as describing an item to sell, a challenge for the participants. In this paper we consider two functions of electronic marketplaces. First, we address the recommendation of related items for users browsing the items offered in a marketplace. Second, in order to support potential sellers we propose the recommendation of relevant items and terms which can be used to describe an item to be sold in the marketplace. The contribution of this paper lies in the proposal of an innovative system that exploits the hidden topics of unstructured information found in the e-marketplace in order to support these functions. We propose a three-step process in which a probabilistic topic modelling approach is used in order to uncover latent topics that provide the basis for item and term similarity calculation for the corresponding recommendations. We present the design of our system and perform evaluations of the quality of the extracted topics as well as of the recommender system using real life scenarios using data extracted from a widely used auction marketplace. The evaluations demonstrate the perceived usefulness of our approach.     

A novel mobile recommender system for indoor shopping

With the widespread usage of mobile terminals, the mobile recommender system is proposed to improve recommendation performance, using positioning technologies. However, due to restrictions of existing positioning technologies, mobile recommender systems are still not being applied to indoor shopping, which continues to be the main shopping mode. In this paper, we develop a mobile recommender system for stores under the circumstance of indoor shopping, based on the proposed novel indoor mobile positioning approach by using received signal patterns of mobile phones, which can overcome the disadvantages of existing positioning technologies. Especially, the mobile recommender system can implicitly capture users’ preferences by analyzing users’ positions, without requiring users’ explicit inputting, and take the contextual information into consideration when making recommendations. A comprehensive experimental evaluation shows the new proposed mobile recommender system achieves much better user satisfaction than the benchmark method, without losing obvious recommendation performances.     

A news-topic recommender system based on keywords extraction

In recent years, internet news has become one of the most important channels for information acquisition, as more and more people read news through internet connected computers, tablets, and smart phones, etc. Owing to the constantly reproduced news, the number of online media increases dramatically and the volume of news also expands rapidly. Consequently, obtaining primary information from the internet is of great interest. This paper presents a news-topic recommender system based on keywords extraction. It is shown that the proposed system is very effective in acquiring specific topics within any specific period of time.     

Applying uncertainty theory into the restaurant recommender system based on sentiment analysis of online Chinese reviews

World Wide Web - In this study, we utilize users’ reviews to a restaurant recommender system to further explore users’ opinions by the proposed recommender approach. Considering the...     

Aspect identification and ratings inference for hotel reviews

Today, a large volume of hotel reviews is available on many websites, such as TripAdvisor (http://www.tripadvisor.com) and Orbitz (http://www.orbitz.com). A typical review contains an overall rating, several aspect ratings, and review text. The rating is an abstract of review in terms of numerical points. The task of aspect-based opinion summarization is to extract aspect-specific opinions hidden in the reviews which do not have aspect ratings, so that users can quickly digest them without actually reading through them. The task consists of aspect identification and aspect rating inference. Most existing studies cannot utilize aspect ratings which become increasingly abundant on review hosts. In this paper, we propose two topic models which explicitly model aspect ratings as observed variables to improve the performance of aspect rating inference on unrated reviews. The experiment results show that our approaches outperform the existing methods on the data set crawled from TripAdvisor website.     

A recommender system to avoid customer churn: A case study

A major concern for modern enterprises is to promote customer value, loyalty and contribution through services such as can help establish a long-term, honest relationship with customers. For purposes of better customer relationship management, data mining technology is commonly used to analyze large quantities of data about customer bargains, purchase preferences, customer churn, etc. This paper aims to propose a recommender system for wireless network companies to understand and avoid customer churn. To ensure the accuracy of the analysis, we use the decision tree algorithm to analyze data of over 60,000 transactions and of more than 4000 members, over a period of three months. The data of the first nine weeks is used as the training data, and that of the last month as the testing data. The results of the experiment are found to be very useful for making strategy recommendations to avoid customer churn.     

Alambic: a privacy-preserving recommender system for electronic commerce

Recommender systems enable merchants to assist customers in finding products that best satisfy their needs. Unfortunately, current recommender systems suffer from various privacy-protection vulnerabilities. Customers should be able to keep private their personal information, including their buying preferences, and they should not be tracked against their will. The commercial interests of merchants should also be protected by allowing them to make accurate recommendations without revealing legitimately compiled valuable information to third parties. We introduce a theoretical approach for a system called Alambic, which achieves the above privacy-protection objectives in a hybrid recommender system that combines content-based, demographic and collaborative filtering techniques. Our system splits customer data between the merchant and a semi-trusted third party, so that neither can derive sensitive information from their share alone. Therefore, the system could only be subverted by a coalition between these two parties.

APPLET: a privacy-preserving framework for location-aware recommender system

Location-aware recommender systems that use location-based ratings to produce recommendations have recently experienced a rapid development and draw significant attention from the research community. However, current work mainly focused on high-quality recommendations while underestimating privacy issues, which can lead to problems of privacy. Such problems are more prominent when service providers, who have limited computational and storage resources, leverage on cloud platforms to fit in with the tremendous number of service requirements and users. In this paper, we propose a novel framework, namely APPLET, for protecting user privacy information, including locations and recommendation results, within a cloud environment. Through this framework, all historical ratings are stored and calculated in ciphertext, allowing us to securely compute the similarities of venues through Paillier encryption, and predict the recommendation results based on Paillier, commutative, and comparable encryption. We also theoretically prove that user information is private and will not be leaked during a recommendation. Finally, empirical results over a real-world dataset demonstrate that our framework can efficiently recommend POIs with a high degree of accuracy in a privacy-preserving manner.     

A multi-agent recommender system for supporting device adaptivity in e-Commerce

Traditional recommender systems for e-Commerce support the customers’ activities providing them with useful suggestions about available products in Web stores. To this purpose, in an agent-based context, each customer is often associated with a customer agent that interacts with the site agent associated with the visited e-Commerce Web site. In presence of a high number of interactions between customers and Web sites, the generation of recommendations can be a heavy task for both these agents. Moreover, customers can navigate on the Web by using different devices having different characteristics that may influence customer’s preferences. In this paper we propose a new multi-agent system, called ARSEC, where each device exploited by a customer is associated with a device agent that autonomously monitors his/her behaviour. Furthermore, each customer is associated with a customer agent that collects in a global profile the information provided by his/her device agents and each e-Commerce Web site is associated with a seller agent. Based on the similarity existing among the global profiles the customers are partitioned in clusters, each one managed by a counsellor agent. Recommendations are generated in ARSEC as result of the collaboration between the seller agent and some counsellor agents associated with the customer. The usage of the device agents leads to generating recommendations taking into account the device currently used, while the fully decentralized architecture introduces a strong reduction of the time costs. Some experimental results are presented to show the significant advantages obtained by ARSEC in terms of recommendation effectiveness with respect to other well-known agent-based recommenders.     

Revyu: Linking reviews and ratings into the Web of Data

Revyu is a live, publicly accessible reviewing and rating Web site, designed to be usable by humans whilst transparently generating machine-readable RDF metadata for the Semantic Web, based on user input. The site uses Semantic Web specifications such as RDF and SPARQL, and the latest Linked Data best practices to create a major node in a potentially Web-wide ecosystem of reviews and related data. Throughout the implementation of Revyu design decisions have been made that aim to minimize the burden on users, by maximizing the reuse of external data sources, and allowing less structured human input (in the form of Web 2.0-style tagging) from which stronger semantics can later be derived. Links to external sources such as DBpedia are exploited to create human-oriented mashups at the HTML level, whilst links are also made in RDF to ensure Revyu plays a first class role in the blossoming Web of Data. In this paper we document design decisions made during the implementation of Revyu, discuss the techniques used for linking Revyu data with external sources, and outline how data from the site is being used to infer the trustworthiness of reviewers as sources of information and recommendations.     

Semi-sparse algorithm based on multi-layer optimization for recommender system

Similarity among vectors is basic knowledge required to carry out recommendation and classification in recommender systems, which support personalized recommendation during online interactions. In this paper, we propose a Semi-sparse Algorithm based on Multi-layer Optimization to speed up the Pearson Correlation Coefficient, which is conventionally used in obtaining similarity among sparse vectors. In accelerating the batch of similarity-comparisons within one thread, the semi-sparse algorithm spares out over-reduplicated accesses and judgements on the selected sparse vector by making this vector dense locally. Moreover, a reduce-vector is proposed to restrict using locks on critical resources in the thread-pool, which is wrapped with Pthreads on a multi-core node to improve parallelism. Furthermore, among processes in our framework, a shared zip file is read to cut down messages within the Message Passing Interface package. Evaluation shows that the optimized multi-layer framework achieves a brilliant speedup on three benchmarks, Netflix, MovieLens and MovieLen1600.