As an effective way to solve information overload, recommender system has drawn attention of scholars from various fields. However, existing works mainly focus on improving the accuracy of... Show More

As an effective way to solve information overload, recommender system has drawn attention of scholars from various fields. However, existing works mainly focus on improving the accuracy of recommendation by designing new algorithms, while the different importance of individual users has not been well addressed. In this paper, the authors propose new approaches to identifying core users based on trust relationships and interest similarity between users, and the popular degree, trust influence and resource of individual users. First, the trust degree and interest similarity between all user pairs, as well as the three attributes of individuals are calculated. Second, a global core user set is constructed based on three strategies, which are frequency-based, rank-based, and fusion-sorting-based. Finally, the authors compare their proposed methods with other existing methods from accuracy, novelty, long-tail distribution and user degree distribution. Experiments show the effectiveness of the authors' core user extraction methods.

With the rapid development of Internet of Things (IoT) and mobile technologies, the service offerings available in the IoT and mobile environments are increasing dramatically. How to provide... Show More

With the rapid development of Internet of Things (IoT) and mobile technologies, the service offerings available in the IoT and mobile environments are increasing dramatically. How to provide intelligent and personalized services for users becomes a challenging issue. Several context aware service recommendation approaches have been reported to leverage roles to represent common knowledge within user communities, based on which services can be recommended for users. Prior studies on context aware role mining mainly focus on mining roles from a fixed data set of user behavior patterns, while most of them neglect the dynamic change of the input data. The frequent change of the user data will result in the change of extracted roles, and how to efficiently update extracted roles according to change of the input user data remains a challenging issue. In this paper, towards this issue, the authors introduce a novel role updating approach in context aware role mining. In the apporach, several algorithms are presented towards various scenarios such as new users and new contexts are removed from and added into the input data. Experiments show that compared with existing solutions, the proposed algorithms can guarantee the completeness of updating results while keeping good updating efficiency.

Nowadays, location based services (LBS) has become one of the most popular applications with the rapid development of mobile Internet technology. More and more research is focused on discovering the... Show More

Nowadays, location based services (LBS) has become one of the most popular applications with the rapid development of mobile Internet technology. More and more research is focused on discovering the required services among massive information according to the personalized behavior. In this paper, a collaborative filtering (CF) recommendation algorithm is presented based on the Location-aware Hidden Markov Model (LHMM). This approach includes three main stages. First, it clusters users by making a pattern similarity calculation of their historical check-in data. Then, it establishes the location-aware transfer matrix so as to get the next most similar service. Furthermore, it integrates the generated LHMM, user's score and interest migration into the traditional CF algorithm so as to generate a final recommendation list. The LHMM-based CF algorithm mixes the geographic factors and personalized behavior and experimental results show that it outperforms the state-of-the-art algorithms on both precision and recall.

Time series classification (TSC) has attracted significant interest over the past decade. A shapelet is one fragment of a time series that can represent class characteristics of the time series. A... Show More

Time series classification (TSC) has attracted significant interest over the past decade. A shapelet is one fragment of a time series that can represent class characteristics of the time series. A classifier based on shapelets is interpretable, more accurate, and faster. However, the time it takes to find shapelets is enormous. This article will propose a fast shapelet (FS) discovery algorithm based on important data points (IDPs). First, the algorithm will identify IDPs. Next, the subsequence containing one or more IDPs will be selected as a candidate shapelet. Finally, the best shapelets will be selected. Results will show that the proposed algorithm reduces the shapelet discovery time by approximately 14.0% while maintaining the same level of classification accuracy rates.

Business processes compliance monitoring checks whether running business processes comply with involved compliance rules. Business processes in modern enterprise are rarely supported by a single and... Show More

Business processes compliance monitoring checks whether running business processes comply with involved compliance rules. Business processes in modern enterprise are rarely supported by a single and centralized workflow system, but instead implemented over different applications (e.g., CRM, ERP, WfMS, and legacy systems). The running data (i.e., event) about process executions are scattered across these applications. Under such circumstance, understanding the compliance of running processes entails the compliance monitoring enabling to correlate events from different applications and even different process instances. This paper introduces a framework named as bpCMon for business process compliance monitoring. bpCMon consists of an expressive compliance rule language ECL and a rule system ERS. ECL is a pattern-based formal language for specifying compliance rules of multiple process perspectives, and also allows for describing event-correlation conditions. ERS, generated from compliance rules in ECL, in turn plays as a compliance monitor enabling to correlate events efficiently by means of an indexing structure created from event-correlation conditions. The applicability of bpCMon is demonstrated by experiments on real-world data sets, and the efficiency of bpCMon is illustrated by comparing with related approaches. Overall, bpCMon enables business process compliance monitoring to meet real-world requirements.