Siyuan Liu

Dynamic witness selection for trustworthy distributed cooperative sensing in cognitive radio networks

Cooperative spectrum sensing by secondary user (SU) nodes in cognitive radio networks (CRNs) is a promising approach to increase the spectrum access efficiency and overall network performance. However, unreliable sensing results or malicious behaviors from cooperator SU nodes can be very disruptive and reduce the network performance. Trust and reputation modeling has been identified as one of the potential solutions to address this problem, but the current centralized trust evaluation approach in CRN lacks scalability. Although some decentralized trust models have been proposed in CRN, without proper protection mechanisms, they are vulnerable to collusive behaviors by the witness SU nodes when they share testimonies about the trustworthiness of neighboring SU nodes. In this paper, we propose a clustering based witness selection method to address this problem. By dividing the witness SU nodes testimonies about the trustworthiness of neighboring SU nodes into clusters, the proposed method helps SU nodes to select which witnesss opinion to trust mode in the future. The proposed method has been studied using extensive computer simulation and has demonstrated good robustness against common collusive attacks.

AN INTEGRATED CLUSTERING-BASED APPROACH TO FILTERING UNFAIR MULTI-NOMINAL TESTIMONIES

Reputation systems have contributed much to the success of electronic marketplaces. However, the problem of unfair testimonies has to be addressed effectively to improve the robustness of reputation systems. Until now, most of the existing approaches focus only on reputation systems using binary testimonies, and thus have limited applicability and effectiveness. In this paper, We propose an integrated CLUstering‐Based approach called iCLUB to filter unfair testimonies for reputation systems using multinominal testimonies, in an example application of multiagent‐based e‐commerce. It adopts clustering techniques and considers buyer agents’ local as well as global knowledge about seller agents. Experimental evaluation demonstrates the promising results of our approach in filtering various types of unfair testimonies, its robustness against collusion attacks, and better performance compared to competing models.

An Incentive Mechanism to Elicit Truthful Opinions for Crowdsourced Multiple Choice Consensus Tasks

Crowdsourcing is a rapidly growing technology to harness human intelligence to solve problems that are not suitable for automation. It is especially suitable for consensus tasks which collect opinions from human workers to gain insight into real-world phenomena. In these tasks, motivating workers to provide their truthful opinions is a challenging problem. Most existing incentive mechanisms proposed to address this problem assume that workers share a common prior. However, this assumption is not always valid in practice, resulting in that the existing approaches may discourage workers to provide their truthful opinions. In this paper, we propose a novel incentive mechanism -- iMET -- to elicit truthful opinions for crowdsourced multiple-choice consensus tasks. By incorporating the concepts of worker credibility and similarity, iMET rewards workers for providing truthful opinions without assuming a common prior. Through extensive simulations on the basis of a collected real-world dataset, iMET has been demonstrated to outperform another two widely used incentive mechanisms in eliciting truthful opinions, especially when diverse truthful opinions are held by workers.

Securing Online Reputation Systems Through Trust Modeling and Temporal Analysis

With the rapid development of reputation systems in various online social networks, manipulations against such systems are evolving quickly. In this paper, we propose scheme TATA, the abbreviation of joint Temporal And Trust Analysis, which protects reputation systems from a new angle: the combination of time domain anomaly detection and Dempster-Shafer theory-based trust computation. Real user attack data collected from a cyber competition is used to construct the testing data set. Compared with two representative reputation schemes and our previous scheme, TATA achieves a significantly better performance in terms of identifying items under attack, detecting malicious users who insert dishonest ratings, and recovering reputation scores.

Using blockchain to build trusted LoRaWAN sharing server

With the rapid growth of the Internet of Things (IoT) market and requirement, low power wide area (LPWA) technologies have become popular. In various LPWA technologies, Narrow Band IoT (NB-IoT) and long range (LoRa) are two main leading competitive technologies. Compared with NB-IoT networks, which are mainly built and managed by mobile network operators, LoRa wide area networks (LoRaWAN) are mainly operated by private companies or organizations, which suggests two issues: trust of the private network operators and lack of network coverage. This study aims to propose a conceptual architecture design of a blockchain built-in solution for LoRaWAN network servers to solve these two issues for LoRaWAN IoT solution.,The study proposed modeling, model analysis and architecture design.,The proposed solution uses the blockchain technology to build an open, trusted, decentralized and tamper-proof system, which provides the indisputable mechanism to verify that the data of a transaction has existed at a specific time in the network.,To the best of our knowledge, this is the first work that integrates blockchain technology and LoRaWAN IoT technology.

A Social Curiosity Inspired Recommendation Model to Improve Precision, Coverage and Diversity

With the prevalence of social networks, social recommendation is rapidly gaining popularity. Currently, social information has mainly been utilized for enhancing rating prediction accuracy, which may not be enough to satisfy user needs. Items with high prediction accuracy tend to be the ones that users are familiar with and may not interest them to explore. In this paper, we take a psychologically inspired view to recommend items that will interest users based on the theory of social curiosity and study its impact on important dimensions of recommender systems. We propose a social curiosity inspired recommendation model which combines both user preferences and user curiosity. The proposed recommendation model is evaluated using large scale real world datasets and the experimental results demonstrate that the inclusion of social curiosity significantly improves recommendation precision, coverage and diversity.

Pumpkin Garden: A Mobile Game Platform for Monitoring Parkinson’s Disease Symptoms

Parkinson’s Disease (PD) is one of the most common neurodegenerative disorders that the elderly are prone to. The recent statistics shows that PD threatens the living quality of over 10 million people worldwide and most of the patients are over 60 ages. Though some medications have been found to be effective in the management of disease progression, the conditions of patients’ symptoms need to be monitored carefully to ensure the effectiveness of appropriate dosage of medications and other necessary treatments to be applied in case that the medications become less effective. Therefore, to facilitate patients and clinicians to have an objective assessment of the conditions of PD symptoms and monitor the effectiveness of treatment, we design a mobile game platform – Pumpkin Garden, which is able to encourage patients to assess their daily conditions through playing games. The patients in-game behaviors are collected and analyzed to generate reports for patients and clinicians to track the response to medications and conditions of disease progression.

Novel Functional Technologies for Age-Friendly E-commerce

In this paper, we present an age-friendly E-commerce system with novel assistive functional technologies, aiming at providing a comfortable online shopping environment for the elderly. Besides incorporating human factors for the elderly into the design of user interface, we build an age-friendly system by improving the functional usability. First, to improve the searching experience, we design a multimodal product search function, which accepts image, speech, text and the combination of them as inputs to help the elderly find products easily and accurately. Second, we develop a product reputation function to provide an objective evaluation of products’ quality, which helps the elderly filter out low-quality products while saves their energy in product comparison. Additionally, to reduce the elderly’s visual burden when browsing the Web, a personalized speech feedback function is designed to provide speech assistant for the elderly. Our system has been testified using real-world E-commerce data, and the result demonstrates its feasibility.