Chunyan Miao

A Survey of Trust and Reputation Management Systems in Wireless Communications

Trust is an important concept in human interactions which facilitates the formation and continued existence of functional human societies. In the first decade of the 21st century, computational trust models have been applied to solve many problems in wireless communication systems. This cross-disciplinary research has yielded many innovative solutions. In this paper, we examine the latest methods which have been proposed by researchers to manage trust and reputation in wireless communication systems. Specifically, we survey the state of the art in the application of trust models in the fields of mobile ad hoc networks (MANETs), wireless sensor networks (WSNs), and cognitive radio networks (CRNs). We classify the mainstream methods into natural categories and illustrate how they complement each other in achieving design goals. Major research directions are also outlined.

Dynamical cognitive network - an extension of fuzzy cognitive map

We present the dynamic cognitive network (DCN) which is an extension of the fuzzy cognitive map (FCM). Each concept in the DCNs can have its own value set, depending on how precisely it needs to be described in the network. This enables the DCN to describe the strength of causes and the degree of effects that are crucial to conducting meaningful inferences. The arcs in the DCN define dynamic, causal relationships between concepts. Structurally, DNCs are scalable and more flexible as compared to FCMs. A DCN can be as simple as a cognitive map and FCM, or as complex as a nonlinear dynamic system. To demonstrate the potential applications of DCNs, we present some simulation results. This paper represents our first attempt to develop a dynamic fuzzy inference system using causal relationships. There are many interesting and challenging theoretical and practical issues in DCNs open to further research.

A Survey of Multi-Agent Trust Management Systems

In open and dynamic multiagent systems (MASs), agents often need to rely on resources or services provided by other agents to accomplish their goals. During this process, agents are exposed to the risk of being exploited by others. These risks, if not mitigated, can cause serious breakdowns in the operation of MASs and threaten their long-term wellbeing. To protect agents from the uncertainty in the behavior of their interaction partners, the age-old mechanism of trust between human beings is re-contexted into MASs. The basic idea is to let agents self-police the MAS by rating each other on the basis of their observed behavior and basing future interaction decisions on such information. Over the past decade, a large number of trust management models were proposed. However, there is a lack of research effort in several key areas, which are critical to the success of trust management in MASs where human beings and agents coexist. The purpose of this paper is to give an overview of existing research in trust management in MASs. We analyze existing trust models from a game theoretic perspective to highlight the special implications of including human beings in an MAS, and propose a possible research agenda to advance the state of the art in this field.

Online multimodal deep similarity learning with application to image retrieval

Recent years have witnessed extensive studies on distance metric learning (DML) for improving similarity search in multimedia information retrieval tasks. Despite their successes, most existing DML methods suffer from two critical limitations: (i) they typically attempt to learn a linear distance function on the input feature space, in which the assumption of linearity limits their capacity of measuring the similarity on complex patterns in real-world applications; (ii) they are often designed for learning distance metrics on uni-modal data, which may not effectively handle the similarity measures for multimedia objects with multimodal representations. To address these limitations, in this paper, we propose a novel framework of online multimodal deep similarity learning (OMDSL), which aims to optimally integrate multiple deep neural networks pretrained with stacked denoising autoencoder. In particular, the proposed framework explores a unified two-stage online learning scheme that consists of (i) learning a flexible nonlinear transformation function for each individual modality, and (ii) learning to find the optimal combination of multiple diverse modalities simultaneously in a coherent process. We conduct an extensive set of experiments to evaluate the performance of the proposed algorithms for multimodal image retrieval tasks, in which the encouraging results validate the effectiveness of the proposed technique.

Towards a trust aware cognitive radio architecture

Cognitive radio (CR) is a promising concept for improving the utilization of scarce radio spectrum resources. A reliable strategy for the detection of unused spectrum bands is essential to the design and practical implementation of CR systems. It is widely accepted that in a real-world environment, cooperative spectrum sensing involving many secondary users scattered in a wide geographical area can greatly improve sensing accuracy. However, some secondary users may misbehave, i.e. provide false sensing information, in an attempt to maximize their own utility gains. Such selfish behaviour, if unchecked, can severely impact the operation of the CR system. In this paper, we propose a novel trustaware hybrid spectrum sensing scheme which can detect misbehaving secondary users and filter out their reported spectrum sensing results from the decision making process. The robustness and efficiency of the proposed scheme are verified through extensive computer simulations.

Implementation of Fuzzy Cognitive Maps Based on Fuzzy Neural Network and Application in Prediction of Time Series

The fuzzy cognitive map (FCM) has gradually emerged as a powerful paradigm for knowledge representation and a simulation mechanism that is applicable to numerous research and application fields. However, since efficient methods to determine the states of the investigated system and to quantify causalities that are the very foundations of FCM theory are lacking, constructing FCMs for complex causal systems greatly depends on expert knowledge. The manually developed models have a substantial shortcoming due to the model subjectivity and difficulties with assessing its reliability. In this paper, we proposed a fuzzy neural network to enhance the learning ability of FCMs. Our approach incorporates the inference mechanism of conventional FCMs with the determination of membership functions, as well as the quantification of causalities. In this manner, FCM models of the investigated systems can automatically be constructed from data and, therefore, operate with less human intervention. In the employed fuzzy neural network, the concept of mutual subsethood is used to describe the causalities, which provides more transparent interpretation for causalities in FCMs. The effectiveness of the proposed approach in handling the prediction of time series is demonstrated through many numerical simulations.

An Entropy-Based Approach to Protecting Rating Systems from Unfair Testimonies

How to mitigate the influence of unfair testimonies remains an open issue in the research of rating systems. Methods have been proposed to filter the unfair testimonies in order to mitigate the influence of unfair testimonies. However, existing methods depend on assumptions that ratings follow a particular distribution to carry out the testimony filtering. This constrains them in specific rating systems and hinders their applications in other reputation systems. Moreover, existing methods do not scale well with the increase of testimony number due to their iterative nature. In this paper, a novel entropy-based method is proposed to measure the testimony quality, based on which unfair testimonies are further filtered. The proposed method does not require the assumption regarding the rating distribution. Moreover, it scales linearly with the increase of the testimony number. Experimental results show that the proposed method is effective in mitigating the influence of various types of unfair testimonies.

Comparing the learning effectiveness of BP, ELM, I-ELM, and SVM for corporate credit ratings

Corporate credit ratings are one of the key problems of the credit risk management, which has attracted much research attention since the credit crisis in 2007. Scorecards are the most widely used approaches for corporate credit ratings nowadays. However, they have heavy dependency on the involvement of users. AI technologies, such as Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs) have demonstrated their remarkable performance on automatic corporate credit ratings. Corporate credit ratings involve various rating models, and their outputs can scale to multiple levels and be used for various applications. Such inherent complexity gives rise to the requirement of higher demands on the effectiveness of learning algorithms regarding the accuracy, overfitness, error distribution, and output distribution. Most research works show that SVMs have better performance than ANNs on accuracy. This paper carries out a comprehensive experimental comparison study over the effectiveness of four learning algorithms, i.e., BP, ELM, I-ELM, and SVM over a data set consisting of real financial data for corporate credit ratings. The results are presented and discussed in the paper.

Challenges and Opportunities for Trust Management in Crowdsourcing

Crowd sourcing (CS) systems offer a new way for businesses and individuals to leverage on the power of mass collaboration to accomplish complex tasks in a divide-and-conquer manner. In existing CS systems, no facility has been provided for analyzing the trustworthiness of workers and providing decision support for allocating tasks to workers, which leads to high dependency of the quality of work on the behavior of workers in CS systems as shown in this paper. To address this problem, trust management mechanisms are urgently needed. Traditional trust management techniques are focused on identifying the most trustworthy service providers (SPs) as accurately as possible. Little thoughts were given to the question of how to utilize these SPs due to two common assumptions: 1) an SP can serve an unlimited number of requests in one time unit, and 2) a service consumer (SC) only needs to select one SP for interaction to complete a task. However, in CS systems, these two assumptions are no longer valid. Thus, existing models cannot be directly used for trust management in CS systems. This paper takes the first step towards a systematic investigation of trust management in CS systems by extending existing trust management models for CS trust management and conducting extensive experiments to study and analyze the performance of various trust management models in crowd sourcing. In this paper, the following key contributions are made. We 1) propose extensions to existing trust management approaches to enable them to operate in CS systems, 2) design a simulation test-bed based on the system characteristics of Amazons Mechanical Turk (AMT) to make evaluation close to practical CS systems, 3) discuss the effect of incorporating trust management into CS system on the overall social welfare, and 4) identify the challenges and opportunities for future trust management research in CS systems.

A new robust training algorithm for a class of single-hidden layer feedforward neural networks

Abstract A robust training algorithm for a class of single-hidden layer feedforward neural networks (SLFNs) with linear nodes and an input tapped-delay-line memory is developed in this paper. It is seen that, in order to remove the effects of the input disturbances and reduce both the structural and empirical risks of the SLFN, the input weights of the SLFN are assigned such that the hidden layer of the SLFN performs as a pre-processor, and the output weights are then trained to minimize the weighted sum of the output error squares as well as the weighted sum of the output weight squares. The performance of an SLFN-based signal classifier trained with the proposed robust algorithm is studied in the simulation section to show the effectiveness and efficiency of the new scheme.