Yi Zhuang

Hybrid Collaborative Filtering algorithm for bidirectional Web service recommendation

Web service recommendation has become a hot yet fundamental research topic in service computing. The most popular technique is the Collaborative Filtering (CF) based on a user-item matrix. However, it cannot well capture the relationship between Web services and providers. To address this issue, we first design a cube model to explicitly describe the relationship among providers, consumers and Web services. And then, we present a Standard Deviation based Hybrid Collaborative Filtering (SD-HCF) for Web Service Recommendation (WSRec) and an Inverse consumer Frequency based User Collaborative Filtering (IF-UCF) for Potential Consumers Recommendation (PCRec). Finally, the decision-making process of bidirectional recommendation is provided for both providers and consumers. Sets of experiments are conducted on real-world data provided by Planet-Lab. In the experiment phase, we show how the parameters of SD-HCF impact on the prediction quality as well as demonstrate that the SD-HCF is much better than extant methods on recommendation quality, including the CF based on user, the CF based on item and general HCF. Experimental comparison between IF-UCF and UCF indicates the effectiveness of adding inverse consumer frequency to UCF.

Alert based disaster notification and resource allocation

When a disaster occurs, timely actions in response to urgent requests conveyed by critical messages (known as alerts) constitute a vital key to effectiveness. These actions include notifying potentially affected parties so that they can take precautionary measures, gathering additional information, and requesting remedial actions and resource allocation. However, there are different types of disasters such as epidemic outbreaks, natural disasters, major accidents, and terrorist attacks. At the same time, there are also many different parties involved such as governments, healthcare institutions, businesses, and individuals. To address these problems, we introduce a Disaster Notification and Resource Allocation System (DNRAS) based on an Alert Management System (AMS) implemented through Web services. This unified platform supports timely interactions among various parties, focusing on notification and monitoring, resource enquiry and allocation, as well as the mobility of information. We detail the mechanisms of these functions in our system, illustrating the Web services interface parameters for communications and interoperability. We illustrate the applicability of our approach with an example of an epidemic outbreak and discuss the advantage of our approach with respect to various stakeholders of our system.

Pick-Up Tree Based Route Recommendation from Taxi Trajectories

Recommending suitable routes to taxi drivers for picking up passengers is helpful to raise their incomes and reduce the gasoline consumption. In this paper, a pick-up tree based route recommender system is proposed to minimize the traveling distance without carrying passengers for a given taxis set. Firstly, we apply clustering approach to the GPS trajectory data of a large number of taxis that indicates state variance from “free” to “occupied”, and take the centroids as potential pick-up points. Secondly, we propose a heuristic based on skyline computation to construct a pick-up tree in which current position is its root node that connects all centroids. Then, we present a probability model to estimate gasoline consumption of every route. By adopting the estimated gasoline consumption as the weight of every route, the weighted Round-Robin recommendation method for the set of taxis is proposed. Our experimental results on real-world taxi trajectories data set have shown that the proposed recommendation method effectively reduce the driving distance before carrying passengers, especially when the number of cabs becomes large. Meanwhile, the time-cost of our method is also lower than the existing methods.

Engineering e-Collaboration Services with a Multi-Agent System Approach

With recent advances in mobile technologies and e-commerce infrastructures, there have been increasing demands for the expansion of collaboration services within and across systems. In particular, human collaboration requirements should be considered together with those for systems and their components. Agent technologies have been deployed in order to model and implement e-commerce activities as multi-agent systems MAS. Agents are able to provide assistance on behalf of their users or systems in collaboration services. As such, we advocate the engineering of e-collaboration support by means of MAS in the following three key dimensions: i across multiple platforms, ii across organization boundaries, and iii agent-based intelligent support. To archive this, we present a MAS infrastructure to facilitate systems and human collaboration or e-collaboration activities based on the belief-desire-intension BDI agent architecture, constraint technology, and contemporary Web Services. Further, the MAS infrastructure also provides users with different options of agent support on different platforms. Motivated by the requirements of mobile professional workforces in large enterprises, the authors present their development and adaptation methodology for e-collaboration services with a case study of constraint-based collaboration protocol from a three-tier implementation architecture aspect. They evaluate our approach from the perspective of three main stakeholders of e-collaboration, which include users, management, and systems developers.

Discovering Communities in Multi-relational Networks

Multi-relational networks (in short as MRNs) refer to such networks including one-typed nodes but associated with each other in poly-relations. MRNs are prevalent in the real world. For example, interactions in social networks include various kinds of information diffusion: email exchange, instant messaging services and so on. Community detection is a long-standing yet very difficult task in social network analysis, especially when meeting MRNs. This chapter gradually explores the research into discovering communities from MRNs. It begins by introducing the generalized modularity of the MRN, which paves the way for applying modularity optimization-based community detection methods on MRNs. However, the mainstream methods for discovering communities on MRNs are to integrate information from multiple dimensions. The existing integration methods fall into four categories: network integration, utility integration, feature integration, and partition integration. Learning or ranking the weight for each relation in MRN constitutes building blocks of network, utility and feature integrations. Thus, we turn our attention into several co-ranking frameworks on MRNs. We then discuss two different kinds of partition integration strategies, including the frequent pattern mining based method and the consensus clustering based method. Finally, for the purpose of conducting performance validation, we present several techniques for constructing the MRN based on both multivariate data and forum data.

A Unified Indexing Structure for Efficient Cross-Media Retrieval

An important trend in web information processing is the support of content-based multimedia retrieval (CBMR). However, the most prevailing paradigm of CBMR, such as content-based image retrieval, content-based audio retrieval, etc, is rather conservative. It can only retrieve media objects of single modality. With the rapid development of Internet, there is a great deal of media objects of different modalities in the multimedia documents such as webpages, which exhibit latent semantic correlation. Cross-media retrieval, as a new multi-media retrieval method, is to retrieve all the related media objects with multi-modalities via submitting a query media object. To the best of our knowledge, this is the first study on how to speed up the cross-media retrieval via indexes. In this paper, based on a Cross-Reference-Graph(CRG )-based similarity retrieval method, we propose a novel unified high-dimensional indexing scheme called CIndex , which is specifically designed to effectively speedup the retrieval performance of the large cross-media databases. In addition, we have conducted comprehensive experiments to testify the effectiveness and efficiency of our proposed method.

A Multi-agent Infrastructure for Healthcare Process Improvement Using Ontology

With the recent rapid advancement in healthcare technologies, there have been increasing expectations in the healthcare services. However, patients are often dissatisfied with the healthcare services and complain through different channels. In this paper, we propose a Multi-Agent Information System (MAIS) framework to help improve healthcare processes with ontology. We illustrate how patient-centered healthcare processes can be facilitated in a large healthcare organization with our framework through a case study in Hong Kong. We also present how ontology can improve operation performance of healthcare processes under our MAIS.

Multiple transmission optimization of medical images in recourse-constraint mobile telemedicine systems

BACKGROUND AND OBJECTIVE In the state-of-the-art image transmission methods, multiple large medical images are usually transmitted one by one which is very inefficient. The objective of our study is to devise an effective and efficient multiple transmission optimization scheme for medical images called Mto via analyzing the visual content of the multiple images based on the characteristics of a recourse-constraint mobile telemedicine system (MTS) and the medical images; METHODS: To better facilitate the efficient Mto processing, two enabling techniques, i.e., 1) NIB grouping scheme, and 2) adaptive RIB replicas selection are developed. Given a set of transmission images (Ω), the correlation of these transmission images is first explored, the pixel resolutions of the corresponding MIBs keep high, the NIBs are grouped into k clusters based on the visual similarity in which the k RIBs are obtained. An optimal pixel resolution for the RIBs is derived based on the current network bandwidth and their corresponding areas, etc. Then, the candidate MIBs and the k RIBs are transmitted to the receiver node based on their transmission priorities. Finally, the IBs are reconstructed and displayed at the receiver node level for different users. RESULTS The experimental results show that our approach is about 45% more efficient than the state-of-the-art methods, significantly minimizing the response time by decreasing the network communication cost while improving the transmission throughput; CONCLUSIONS: Our proposed Mto method can be seamlessly applied in a recourse-constraint MTS environment in which the high transmission efficiency and the acceptable image quality can be guaranteed.

A Personalized Travel System Based on Crowdsourcing Model

With the proliferation of the online tourism markets, and the rapid change of tourists demands, existing online travel platforms cannot satisfy tourists to some extent, since their tourism demands tend to be more personalized and dynamic. Based on the above motivations, we design and develop a personalized tourism system based on a novel cooperation crowdsourcing model through the Internet. More importantly, data quality control based on the crowdsourcing model is a key problem which affects the accuracy and effectiveness of tourist recommendation. To address this problem, we propose three data quality control schemes for personalized tours based on the crowdsourcing model. Extensive experiments validate the effectiveness of our proposed approach.

A hybrid-feature-based efficient retrieval over Chinese calligraphic manuscript image repository

In this paper, we propose an interactive indexing scheme to facilitate an efficient search of Chinese calligraphic manuscript images based on hybrid features such as contour points and number of strokes. Different from conventional character retrieval and indexing methods [6], our proposed indexing algorithm allows user to choose other feature such as stroke of character they prefer to as query element. To better facilitate an efficient and unified retrieval, we then propose a Hybrid-Feature-Tree(HF-Tree)-based high- dimensional indexing scheme. Comprehensive experiments are conducted to testify its effectiveness and efficiency.