TRUST BY ASSOCIATION: A META‐REPUTATION SYSTEM FOR PEER‐TO‐PEER NETWORKS

Trust mechanisms are used in peer‐to‐peer (P2P) networks to help well‐behaving peers find other well‐behaving peers with which to trade. Unfortunately, these trust mechanisms often do little to keep badly behaving peers from entering and taking advantage of the network, which makes the resulting network difficult or impossible to use for legitimate purposes such as e‐commerce. We propose trust by association, a way of tying peers together in invitation‐only P2P networks in such a way as to encourage the removal of badly behaving peers. We use invitations to create a structure within the otherwise ad hoc P2P network. Using this structure, we create a meta‐reputation system where we measure a peer’s trustworthiness not only by its own behavior, but also by the behavior of the peers it has invited to join. The connection created between the peers takes advantage of the external social relationship that must exist before a peer can be invited into the network. The result is a P2P network where, rather than just trying to marginalize badly behaving peers, there is incentive to kick them out of the network. We present results from a simple simulation showing that our approach works well in general when combined with and compared to an existing trust mechanism.     

A comprehensive peer‐to‐peer characterization model

A peer‐to‐peer (P2P) characterization model is proposed in this paper. The purposes of this model are (a) to describe through abstraction the behavior and capability of P2P computing independent of any specific application, and (b) to lead to a cost model of P2P computing. Our approach is based on abstracting the service driven nature of P2P computing separate and distinct from its implementation. This supports the first purpose. Next, we consider the integration of the two aspects. This supports the second purpose. A secondary contribution of our work is a cost model. The cost model is applied to indexing, discovery and resource utilization services. In addition, the characterization model and cost model are applied to a JXTA‐based P2P application. Copyright © 2006 John Wiley & Sons, Ltd.     

Zorilla: a peer‐to‐peer middleware for real‐world distributed systems

The inherent complex nature of current distributed computing architectures hinders the widespread adoption of these systems for mainstream use. In general, users have access to a highly heterogeneous set of compute resources, which may include clusters, grids, desktop grids, clouds, and other compute platforms. This heterogeneity is especially problematic when running parallel and distributed applications. Software is needed which easily combines as many resources as possible into one coherent computing platform. In this paper, we introduce Zorilla: peer‐to‐peer (P2P) middleware that creates a single distributed environment from any available set of compute resources. Zorilla imposes minimal requirements on the resource used, is platform independent, and does not rely on central components. In addition to providing functionality on bare resources, Zorilla can exploit locally available middleware. Zorilla explicitly supports distributed and parallel applications, and allows resources from multiple sites to cooperate in a single computation. Zorilla makes extensive use of both virtualization and P2P techniques. We will demonstrate how virtualization and P2P combine into a simple design, while enhancing functionality and ease of use. Together, these techniques bring our goal a step closer: transparent, easy use of resources, even on very heterogeneous distributed systems. Copyright © 2011 John Wiley & Sons, Ltd.     

Predicting repayment of borrows in peer‐to‐peer social lending with deep dense convolutional network

In peer‐to‐peer lending, it is important to predict the repayment of the borrower to reduce the lender's financial loss. However, it is difficult to design a powerful feature extractor for predicting the repayment as user and transaction data continue to increase. Convolutional neural networks automatically extract useful features from big data, but they use only high‐level features; hence, it is difficult to capture a variety of representations. In this study, we propose a deep dense convolutional network for repayment prediction in social lending, which maintains the borrower's semantic information and obtains a good representation by automatically extracting important low‐ and high‐level features simultaneously. We predict the repayment of the borrower by learning discriminative features depending on the loan status. Experimental results on the Lending Club dataset show that our model is more effective than other methods. A fivefold cross‐validation is performed to run the experiments.     

The problem of upload competition in peer‐to‐peer systems with incentive mechanisms

As peer‐to‐peer (P2P) file‐sharing systems revolve around cooperation, the design of upload incentives has been one of the most important topics in P2P research for more than a decade. Several deployed systems, such as private BitTorrent communities, successfully manage to foster cooperation by banning peers when their sharing ratio becomes too low. Interestingly, recent measurements have shown that such systems tend to have an oversupply instead of an undersupply of bandwidth designers that have been obsessed with since the dawn of P2P. In such systems, the ‘selfish peer’ problem is finally solved, but a new problem has arisen: because peers have to keep up their sharing ratios, they now have to compete to upload. In this paper, we explore this new problem and show how even highly cooperative peers might in the end not survive the upload competition. On the basis of recent measurements of over half a million peers in private P2P communities, we propose and analyze several algorithms for uploader selection under oversupply. Our algorithms enable sustained sharing ratio enforcement and are easy to implement in both existing and new systems. Overall, we offer an important design consideration for the new generation of P2P systems in which selfishness is no longer an issue. Copyright © 2012 John Wiley & Sons, Ltd.     

Comparison of individual,ensemble and integrated ensemble machine learning methods to predict China’s SME credit risk in supply chain finance

Supply chain finance (SCF) becomes more important for small- and medium-sized enterprises (SMEs) due to global credit crunch, supply chain financing woes and tightening credit criteria for corporate lending. Currently, predicting SME credit risk is significant for guaranteeing SCF in smooth operation. In this paper, we apply six methods, i.e., one individual machine learning (IML, i.e., decision tree) method, three ensemble machine learning methods [EML, i.e., bagging, boosting, and random subspace (RS)], and two integrated ensemble machine learning methods (IEML, i.e., RS–boosting and multi-boosting), to predict SMEs credit risk in SCF and compare the effectiveness and feasibility of six methods. In the experiment, we choose the quarterly financial and non-financial data of 48 listed SMEs from Small and Medium Enterprise Board of Shenzhen Stock Exchange, six listed core enterprises (CEs) from Shanghai Stock Exchange and three listed CEs from Shenzhen Stock Exchange during the period of 2012–2013 as the empirical samples. Experimental results reveal that the IEML methods acquire better performance than IML and EML method. In particular, RS–boosting is the best method to predict SMEs credit risk among six methods.

     

A flexible content repository to enable a peer‐to‐peer‐based wiki

Wikis—being major applications of the Web 2.0—are used for a large number of purposes, such as encyclopedias, project documentation, and coordination, both in open communities and in enterprises. At the application level, users are targeted as both consumers and producers of dynamic content. Yet, this kind of peer‐to‐peer (P2P) principle is not used at the technical level being still dominated by traditional client–server architectures. What lacks is a generic platform that combines the scalability of the P2P approach with, for example, a wiki's requirements for consistent content management in a highly concurrent environment. This paper presents a flexible content repository system that is intended to close the gap by using a hybrid P2P overlay to support scalable, fault‐tolerant, consistent, and efficient data operations for the dynamic content of wikis. On the one hand, this paper introduces the generic, overall architecture of the content repository. On the other hand, it describes the major building blocks to enable P2P data management at the system's persistent storage layer, and how these may be used to implement a P2P‐based wiki application: (i) a P2P back‐end administrates a wiki's actual content resources. (ii) On top, P2P service groups act as indexing groups to implement a wiki's search index. Copyright © 2009 John Wiley & Sons, Ltd.     

Efficient resource discovery in self‐organized unstructured peer‐to‐peer networks

In unstructured peer‐to‐peer (P2P) networks, two autonomous peer nodes can be connected if users in those nodes are interested in each other's data. Owing to the similarity between P2P networks and social networks, where peer nodes can be regarded as people and connections can be regarded as relationships, social strategies are useful for improving the performance of resource discovery by self‐organizing autonomous peers on unstructured P2P networks. In this paper, we present an efficient social‐like peer‐to‐peer (ESLP) method for resource discovery by mimicking different human behaviours in social networks. ESLP has been simulated in a dynamic environment with a growing number of peer nodes. From the simulation results and analysis, ESLP achieved better performance than current methods. Copyright © 2008 John Wiley & Sons, Ltd.     

A comparative study of awareness methods for peer‐to‐peer distributed virtual environments

The increasing popularity of multi‐player online games is leading to the widespread use of large‐scale Distributed Virtual Environments (DVEs) nowadays. In these systems, peer‐to‐peer (P2P) architectures have been proposed as an efficient and scalable solution for supporting massively multi‐player applications. However, the main challenge for P2P architectures consists of providing each avatar with updated information about which other avatars are its neighbors. This problem is known as the awareness problem. In this paper, we propose a comparative study of the performance provided by those awareness methods that are supposed to fully solve the awareness problem. This study is performed using well‐known performance metrics in distributed systems. Moreover, while the evaluations shown in the literature are performed by executing P2P simulations on a single (sequential) computer, this paper evaluates the performance of the considered methods on actually distributed systems. The evaluation results show that only a single method actually provides full awareness to avatars. This method also provides the best performance results. Copyright © 2008 John Wiley & Sons, Ltd.     

Financing decisions in supply chains with a capital‐constrained manufacturer: competition and risk

Supply chain finance is a crucial topic. In this paper, we consider that a capital‐constrained manufacturer can borrow money from either a bank (bank credit financing) or a retailer (trade credit financing). Our analysis compares supply chain performance under these two financing schemes. Furthermore, we extend our model to evaluate the impacts of retail competition and supply chain member's risk aversion on supply chains, which consist of one capital‐constrained manufacturer and two competing retailers. We consider three financing schemes: only bank credit financing, dual trade credit financing, and bank and trade credit mix financing. We find that without retail competition, the retailer is always willing to use the trade credit financing; whereas with retail competition, if one retailer provides the trade credit but the other does not, the credit provider could receive the superior profit. Thus, providing an appropriate trade credit financing scheme is critically important for retailers. Moreover, we find that without retail competition, when a trade interest rate is relatively low, both the retailer and manufacturer could reach a win‐win situation in the trade credit financing. However, with retail competition, supply chain members (i.e., two retailers and one manufacturer) will not have an all‐win situation no matter which specific financing scheme is adopted and only a win‐win‐lose situation exists when using the credit mix financing scheme or the dual trade credit financing in supply chains. Last but not least, regardless of risk neutrality or aversion of supply chain members, their pricing decisions among three financing schemes are similar. This implies that the impacts of supply chain members’ risk aversion are limited in supply chain financing scheme selection. More managerial insights are discussed.     

Analysis and design of peer‐assisted video‐on‐demand services

We propose a deterministic fluid model to understand the trade‐offs in the design of peer‐assisted video‐on‐demand (VoD) services. There are three entities in this model: peers (or end users), seeders (altruistic users that own one or several complete video items), and cache servers that store and forward videos with a limited capacity. Peers join the network, download one or multiple concurrent video streams (possibly different video items), and abort the system when they wish. Peers are assumed to cooperate in a BitTorrent‐based fashion, governed by tit‐for‐tat and fair availability. The issue is to minimize the expected downloading times, choosing the set of video items that should be stored in each cache server. We first prove that the deterministic model is globally stable, and find closed expressions for the expected waiting times. Then, we introduce a combinatorial optimization problem (COP), whose nature is similar to the multiknapsack problem (where the items are videos, and knapsacks are related with the storage of cache servers). The problem turns out to be NP‐complete. Therefore, we heuristically address the problem following a GRASP (greedy randomized adaptive search procedure) methodology. Finally, we simulate the new caching methodology with real‐life traces taken from YouTube logs. The results suggest that the peer‐to‐peer philosophy is both stable and cost‐effective for on‐demand streaming purposes.     

一种支持多维资源描述的高效P2P路由算法

在分析现有P2P(peer to peer)路由算法的基础上,提出了一种基于二阶矩定位、支持多维资源数据描述的高效资源路由算法--FAN(flabellate addressable network)路由算法.FAN算法将节点映射到统一的多维笛卡尔空间,并以节点相对空间原点的二阶矩作为子空间管理和资源搜索的依据.FAN路由算法具有O(log(N/k))的高路由效率,在节点加入和退出FAN网络时,更新路由信息的代价为O(klog(N/k)).实验结果表明,FAN路由算法具有路由效率高、维护代价小的优点,是一种P2P环境中支持多维资源数据描述的高效结构化资源路由算法.而且,目前部分基于CAN(content-addressable network)网络的改进算法也可以在FAN网络中适用,并获得更好的路由效率和更低的维护代价.     

The impact of team‐based learning on students with different self‐regulated learning abilities

Team‐based learning (TBL) stresses applying knowledge rather than absorbing knowledge in class; studies have investigated the use of TBL and its merits in different teaching courses (e.g., medical science and business). TBL is most effective when students learn autonomously before class. However, the ability of autonomous learning is highly associated with the ability of self‐regulated learning (SRL); most importantly, not every student possesses good (or high) SRL ability. Nevertheless, few studies have compared the effectiveness of TBL in students with different SRL abilities. To address this issue, this study analyzed approximately 90 students, whose course teaching involves office application software (Microsoft Excel). This study also developed an online TBL system (called Online TBL) to facilitate performing TBL and to collect the learning behaviours of students with different (high or low) SRL abilities on each TBL stage. The analytical results show that compared with the low‐SRL students, the high‐SRL students were more prepared for class because they spent more reviewing material and had better scores on personal uploaded Excel and Individual Readiness Assurance Test. From the feedback of the peer evaluation, the results also show that the high‐SRL students received more credits than the low‐SRL students did. The questionnaire survey revealed that both low‐SRL and high‐SRL students had a favourable impression of TBL. Further discussion is given to explain the above results.     

Pay‐as‐you‐throw versus recycling fund system in closed‐loop supply chains with alliance recycling

This paper develops closed‐loop supply chain models with alliance recycling under two financing systems for recycling used products, that is, pay‐as‐you‐throw (PAYT) and recycling fund (RF). The primary goal behind analytic formulation is to investigate the differences between the two financing systems and determine which one performs better, and to examine the impact of target recycling rate and competition between recycling alliances (RAs) on channel decisions. We find that the PAYT system is always more beneficial to customers and the retailer, while the RF system increases the social welfare. If there is no competition between RAs, which system performs better for the RA depends on the baseline of subsidy fee, while which system performs better for manufacturers depends on their shares. However, if there is competition between RAs, the RF system is favored by the disadvantaged RA. Moreover, the restriction on target recycling rate has an effect on channel decisions.     

A database approach for information communication in a peer‐to‐peer collaborative CAD environment

Timely and efficient information communication is a key factor in ensuring successful collaboration in engineering collaborative design. This work proposes a database approach to support information communication between distributed and autonomous CAD systems. It provides the designer with an easy and flexible way, a project‐based propagation meta‐table, to specify what parts of a CAD information model should be communicated to other collaborating designers. A CAD peer manager, containing a peer database that stores information to be exchanged with the other collaborators, wraps each participating CAD system. The peer manager identifies changes made to the CAD model by using stored procedures and active rules in the peer database that are automatically generated based on the propagation meta‐table. The identified updates are propagated in a timely manner to other peers via inter‐database message passing, thereby minimizing the volume of necessary information to be exchanged. Furthermore, remote peer designers can flexibly incorporate, filter, or delete received updates by using a propagation control interface, which is also used to issue user's commands to download the data from the CAD system to the peer database and lookup the received messages in the peer database. The approach is applicable on any CAD system having a CORBA interface and can also be applied to other kinds of object‐oriented interfaces. Copyright © 2007 John Wiley & Sons, Ltd.     

Myriad: scalable VR via peer‐to‐peer connectivity,PC clustering,and transient inconsistency

Distributed scene graphs are important in virtual reality, both in collaborative virtual environments and in cluster rendering. Modern scalable visualization systems have high local throughput, but collaborative virtual environments (VEs) over a wide‐area network (WAN) share data at much lower rates. This complicates the use of one scene graph across the whole application. Myriad is an extension of the Syzygy VR toolkit in which individual scene graphs form a peer‐to‐peer network. Myriad connections filter scene graph updates and create flexible relationships between nodes of the scene graph. Myriad's sharing is fine‐grained: the properties of individual scene graph nodes to share are dynamically specified (in C++ or Python). Myriad permits transient inconsistency, relaxing resource requirements in collaborative VEs. A test application, WorldWideCrowd, demonstrates collaborative prototyping of a 300‐avatar crowd animation viewed on two PC‐cluster displays and edited on low‐powered laptops, desktops, and over a WAN. We have further used our framework to facilitate collaborative educational experiences and as a vehicle for undergraduates to experiment with shared virtual worlds. Copyright © 2006 John Wiley & Sons, Ltd.     

JXTA下P2P点组认证协议的设计和实现

JXTA是由SUN公司推出的一项旨在为P2P应用而建立的一个通用开发平台。虽然JXTA平台提供了成员协议来验证Peer加入点组,但它没有提供认证要求的修改,在申请过程中也没有组内成员的参与。该文扩展了一种基于投票的、灵活的点组认证协议,通过实例对点组认证协议的可行性、正确性进行了验证。     

Fostering critical thinking and reflection through blog‐mediated peer feedback

The introduction of digital literacy practices has created a tension in academia, with many academics challenging the view that critical thinking can be fostered on social networks. A quasi‐experimental study was conducted on two sections of university‐level writing classrooms to determine if there were meaningful differences in the quality of writing and peer feedback generated through in‐class draft workshops using traditional methods as compared to draft workshops using a blogging platform. The results indicated that blogs produced a higher quality of writing as measured by grades, f(42) = 11.512, p < .002 and acceptance scores, f(42) = 8.364, p < .006. Furthermore, blog‐mediated peer workshops produced a statistically significantly higher number of critical comments, f(42) = 120.438, p < .000; and directive comments, f(42) = 33.861, p < .000. There were no statistically significant differences in the number of editing comments, f(42) = .001, p < .974, and traditional draft workshops produced a statistically significant higher number of naïve comments, f(42) = 14.119, p < .001. Within the study, critical comments were found to correlate positively with learning outcomes, b = 1.115, t(41) = 2.716, p < .01. The findings suggest that blogging software improved learner performance and fostered complex literacy skills.     

Topology,typology, and dynamics of commons‐based peer production: On platforms,actors, and innovation in the maker movement

In this paper, we aim to present a comprehensive analysis on the emerging phenomenon of distributed innovation in commons‐based peer production (CBPP) platforms. Starting with the exploration of the widely held belief on value‐creation confined to industrial settings, we raise several questions regarding the feasibility of, and a need for, an inclusive innovation process that can tap grassroots capacity into both traditional (industrial research and development) and emerging (peer‐to‐peer) innovation models to yield sustainable solutions. In particular, we explore the emergence and structuration of digital innovations in the maker movement, as it presents an alternative construct of innovation wherein access to and sharing of knowledge is predominantly distributed. With innovation outcomes often freely revealed, its very structuration could pose a principal challenge to our conceptualizations of value creation and competitive advantage in the current economic model. Drawing from responses received from 200 collaborative innovation platforms, six semi‐structured interviews focusing on socio‐technical innovation cases, as well as four in‐depth narrative interviews with maker turned entrepreneurs, we present a detailed analysis on the topology of network, typology of actors, as well as the underlying innovation ecosystem in CBPP networks in Germany. In doing so, we contribute to the conceptualization of peer‐to‐peer distributed innovations in collaborative platforms.     

Just‐in‐Time Modeling for Function Prediction and Its Applications

This paper introduces a variant of k Bipartite Neighbors (k‐BN), called k‐BN2, for use in function prediction. Like k‐BN, k‐BN2 selects k instances surrounding the query, i.e., the novel instance, and keeps them bipartitely. However, in order to improve the prediction precision, based on the bipartite neighborhood, k‐BN2 combines local linear models and the global nonlinear model to predict the value of the novel instance. Applied to two real measured datasets, k‐BN2 outperforms the typical k‐BN andthose methods in which k‐BN or a related approximate physical model alone is exploited.