Donghai Guan

Improved trust-aware recommender system using small-worldness of trust networks

The trust network is a social network where nodes are inter-linked by their trust relations. It has been widely used in various applications, however, little is known about its structure due to its highly dynamic nature. Based on five trust networks obtained from the real online sites, we contribute to verify that the trust network is the small-world network: the nodes are highly clustered, while the distance between two randomly selected nodes is short. This has considerable implications on using the trust network in the trust-aware applications. We choose the trust-aware recommender system as an example of such applications and demonstrate its advantages by making use of our verified small-world nature of the trust network.

Activity Recognition Based on Semi-supervised Learning

Activity recognition is a hot topic in context-aware computing. In activity recognition, machine learning techniques have been widely applied to learn the activity models from labeled activity samples. Since labeling samples requires humans efforts, most existing research in activity recognition focus on refining learning techniques to utilize the costly labeled samples as effectively as possible. However, few of them consider using the costless unlabeled samples to boost learning performance. In this work, we propose a novel semi-supervised learning algorithm named En-Co-training to make use of the unlabeled samples. Our algorithm extends the co- training paradigm by using ensemble method. Experimental results show that En-Co-training is able to utilize the available unlabeled samples to enhance the performance of activity learning with a limited number of labeled samples.

ITARS: trust-aware recommender system using implicit trust networks

Trust-aware recommender system (TARS) suggests the worthwhile information to the users on the basis of trust. Existing works of TARS suffers from the problem that they need extra user efforts to label the trust statements. The authors propose a novel model named iTARS to improve the existing TARS by using the implicit trust networks: instead of using the effort-consuming explicit trust, the easy available user similarity information is used to generate the implicit trusts for TARS. Further analysis shows that the implicit trust network has the small-world topology, which is independent of its dynamics. The rating prediction mechanism of iTARS is based on the small worldness of the implicit trust network: the authors set the maximum trust propagation distance of iTARS approximately equals the average path length of the trust networks corresponding random network. Experimental results show that with the same computational complexity, iTARS is able to improve the existing TARS works with higher rating prediction accuracy and slightly worse rating prediction coverage.

The small-world trust network

The topology of the trust network is important to optimize its usage in the trust-aware applications. However, since the users can join trust network ubiquitously, the structure of the highly dynamic trust network is still unknown. This paper contributes to verify that the trust network is the small-world network, and its small-world topology is independent of its dynamics. This is achieved by verifying the scale-freeness of five trust networks extracted from real online sites. Using the small-world nature of the trust network, we optimize the rating prediction mechanism of the conventional trust-aware recommender system. Experimental results clearly show that our proposed mechanism can achieve the maximum accuracy and coverage with the minimum computation complexity for the rating predictions.

Nearest neighbor editing aided by unlabeled data

This paper proposes a novel method for nearest neighbor editing. Nearest neighbor editing aims to increase the classifiers generalization ability by removing noisy instances from the training set. Traditionally nearest neighbor editing edits (removes/retains) each instance by the voting of the instances in the training set (labeled instances). However, motivated by semi-supervised learning, we propose a novel editing methodology which edits each training instance by the voting of all the available instances (both labeled and unlabeled instances). We expect that the editing performance could be boosted by appropriately using unlabeled data. Our idea relies on the fact that in many applications, in addition to the training instances, many unlabeled instances are also available since they do not need human annotation effort. Three popular data editing methods, including edited nearest neighbor, repeated edited nearest neighbor and All k-NN are adopted to verify our idea. They are tested on a set of UCI data sets. Experimental results indicate that all the three editing methods can achieve improved performance with the aid of unlabeled data. Moreover, the improvement is more remarkable when the ratio of training data to unlabeled data is small.

Identifying mislabeled training data with the aid of unlabeled data

This paper presents a new approach for identifying and eliminating mislabeled training instances for supervised learning algorithms. The novelty of this approach lies in the using of unlabeled instances to aid the detection of mislabeled training instances. This is in contrast with existing methods which rely upon only the labeled training instances. Our approach is straightforward and can be applied to many existing noise detection methods with only marginal modifications on them as required. To assess the benefit of our approach, we choose two popular noise detection methods: majority filtering (MF) and consensus filtering (CF). MFAUD/CFAUD is the new proposed variant of MF/CF which relies on our approach and denotes majority/consensus filtering with the aid of unlabeled data. Empirical study validates the superiority of our approach and shows that MFAUD and CFAUD can significantly improve the performances of MF and CF under different noise ratios and labeled ratios. In addition, the improvement is more remarkable when the noise ratio is greater.

Review of Sensor-based Activity Recognition Systems

Abstract Activity recognition (AR) has become a hot research topic due to its strength in providing personalized sup port for many diverse applications such as healthcare and security. Due to its importance, a considerable amount of AR systems have been developed. In general, these systems utilize diverse sensors to obtain the activity related information, which are then used by machine learning techniques to infer human’s ongoing activity. According to the types of sensors used, existing AR systems can be roughly divided into two catego ries: 1. Video sensor based AR. It remotely observes human activity using video sensors; 2. Physical sensor based AR (PSAR). It attaches physical sensors to the body of human or objects (appliances) to infer human activity. Based on the location of attached sensors, PSAR consists of two subcategories: Wearable sensor based AR and object usage based AR. In this work, different types of AR are reviewed. We think this review is significant because no existing review papers include all the different types of AR as a whole. For each type of AR, its main techniques, characteristics, strengths and limitations are discussed and summarized. We also make a comparative analysis for them. Finally the main research challenges in AR are pointed out.

Towards Smart Homes Using Low Level Sensory Data

Ubiquitous Life Care (u-Life care) is receiving attention because it provides high quality and low cost care services. To provide spontaneous and robust healthcare services, knowledge of a patient’s real-time daily life activities is required. Context information with real-time daily life activities can help to provide better services and to improve healthcare delivery. The performance and accuracy of existing life care systems is not reliable, even with a limited number of services. This paper presents a Human Activity Recognition Engine (HARE) that monitors human health as well as activities using heterogeneous sensor technology and processes these activities intelligently on a Cloud platform for providing improved care at low cost. We focus on activity recognition using video-based, wearable sensor-based, and location-based activity recognition engines and then use intelligent processing to analyze the context of the activities performed. The experimental results of all the components showed good accuracy against existing techniques. The system is deployed on Cloud for Alzheimer’s disease patients (as a case study) with four activity recognition engines to identify low level activity from the raw data captured by sensors. These are then manipulated using ontology to infer higher level activities and make decisions about a patient’s activity using patient profile information and customized rules.

Grid Task Scheduling: Algorithm Review

Abstract As a new distributed heterogeneous computing platform, grid aims at achieving Internet-wide resource sharing and collaborative computing. Grid task scheduling (GTS) is the key issue of grid computing, and its algorithm has a direct effect on the performance of the whole system. In this paper, two key entities in GTS, applications and target systems, are defined first. And then two types of the most popular GTS algorithms, namely, meta-task GTS algorithm and directed acyclic graph GTS algorithm, are discussed in details in accordance with the classification of the traditional deterministic algorithm and heuristic intelligent algorithm. In addition, the comparative analysis is made among them. Finally, some main research directions of GTS are pointed out.

Context-aware Human Activity Recognition and decision making

Ubiquitous Life Care (u-Life care) nowadays becomes more attractive to computer science researchers due to a demand on a high quality and low cost of care services at anytime and anywhere. Many works exploit sensor networks to monitor patients health status, movements, and real-time daily life activities to provide care services to them. Context information with real-time daily life activities can help in better services, service suggestions, and change in system behavior for better healthcare. Our proposed Secured Wireless Sensor Network - integrated Cloud Computing for ubiquitous - Life Care (SC3) monitors human health as well as activities. In this paper we focus on Human Activity Recognition Engine (HARE) framework architecture, backbone of SC3 and discussed it in detail. Camera-based and sensor-based activity recognition engines are discussed in detail along with the manipulation of recognized activities using Context-aware Activity Manipulation Engine (CAME) and Intelligent Life Style Provider (i-LiSP). Preliminary results of CAME showed robust and accurate response to medical emergencies. We have deployed five different activity recognition engines on Cloud to identify different set of activities of Alzheimers disease patients.