The value of big data for credit scoring: Enhancing financial inclusion using mobile phone data and social network analytics

Credit scoring is without a doubt one of the oldest applications of analytics. In recent years, a multitude of sophisticated classification techniques have been developed to improve the statistical performance of credit scoring models. Instead of focusing on the techniques themselves, this paper leverages alternative data sources to enhance both statistical and economic model performance. The study demonstrates how including call networks, in the context of positive credit information, as a new Big Data source has added value in terms of profit by applying a profit measure and profit-based feature selection. A unique combination of datasets, including call-detail records, credit and debit account information of customers is used to create scorecards for credit card applicants. Call-detail records are used to build call networks and advanced social network analytics techniques are applied to propagate influence from prior defaulters throughout the network to produce influence scores. The results show that combining call-detail records with traditional data in credit scoring models significantly increases their performance when measured in AUC. In terms of profit, the best model is the one built with only calling behavior features. In addition, the calling behavior features are the most predictive in other models, both in terms of statistical and economic performance. The results have an impact in terms of ethical use of call-detail records, regulatory implications, financial inclusion, as well as data sharing and privacy.     

Analytical framework for dimensioning hierarchical WiMax–WiFi networks

Providing diverse, ubiquitous and cost-effective broadband services is a foremost challenge for the telecommunication community. Fixed WiMAX or IEEE 802.16d is one of the most promising radio access technologies, providing high performance similar to wired xDSL systems, yet superior to that of current 3G mobile technologies. Numerous deployment concepts are foreseen for WiMAX networks. They are designed to cover isolated areas, thus embodying an appealing solution for cellular networks or wireless backhaul for WiFi access. The latter concept is of interest in this paper that puts forward an analytical model based on the economical aspects to dimension hierarchical WiMax–WiFi networks. The proposed model consists in replacing a finite number of nodes by an equivalent continuum. Its key feature lies in accounting for the effect of interference as well as for the physical layer and channel characteristics in an easy and straightforward manner. On the one hand, the model takes into consideration frequency planning and scheduling aspects; and on the other hand, it provides tractable formulae of the end-user mean capacity and coverage probability in order to properly dimension the hybrid network. Last but not least, the economical facet of network planning is considered to unravel the design trade-offs between maximizing the service provider profit and satisfying the end user requirements in terms of performance.     

Toward cost-sensitive self-optimizing anomaly detection and response in autonomic networks

While anomaly detection and response play a significant role in attaining auto defense, one of core functionalities of autonomic networks, the design and deployment of Anomaly Detection and Response Systems (ADRS) herein is a non-trivial issue because of the special network characteristic, namely self-managing, which requires candidate ADRS to automatically and optimally balance performance objectives and potential negative consequence. In this paper, we propose a decision-theoretic framework to systematically analyze ADRS in autonomic networks, with an objective to achieve its cost-sensitive and self-optimizing operation. In particular, each ADRS agent is viewed as an autonomous entity, making decision as its local operating environment. A global reward signal is then used to quantify the performance of ADRS as a whole in terms of those identified metrics. Furthermore, the analytical framework serves as a basis for developing an adaptive, robust, and near-optimal prototype termed ARSoS, along with a reinforcement learning algorithm for approximately inferring the optimal behavior of a reputation-based ADRS in a specific autonomic network variant, mobile ad-hoc network. The performance of ARSoS is validated through extensive simulations.     

Dynamic Ridge Polynomial Neural Network: Forecasting the univariate non-stationary and stationary trading signals

This paper considers the prediction of noisy time series data, specifically, the prediction of financial signals. A novel Dynamic Ridge Polynomial Neural Network (DRPNN) for financial time series prediction is presented which combines the properties of both higher order and recurrent neural network. In an attempt to overcome the stability and convergence problems in the proposed DRPNN, the stability convergence of DRPNN is derived to ensure that the network posses a unique equilibrium state. In order to provide a more accurate comparative evaluation in terms of profit earning, empirical testing used in this work encompass not only on the more traditional criteria of NMSE, which concerned at how good the forecasts fit their target, but also on financial metrics where the objective is to use the networks predictions to generate profit. Extensive simulations for the prediction of one and five steps ahead of stationary and non-stationary time series were performed. The resulting forecast made by DRPNN shows substantial profits on financial historical signals when compared to various neural networks; the Pi-Sigma Neural Network, the Functional Link Neural Network, the feedforward Ridge Polynomial Neural Network, and the Multilayer Perceptron. Simulation results indicate that DRPNN in most cases demonstrated advantages in capturing chaotic movement in the financial signals with an improvement in the profit return and rapid convergence over other network models.     

深度神经架构搜索综述

深度学习在图像、语音、文本等多种模态的数据任务上取得了优异的效果.然而,针对特定任务,人工设计网络需要花费大量的时间,并且需要设计者具有一定水平的专业知识和设计经验.面对如今日趋复杂的网络架构,仅依靠人工进行设计变得越来越复杂.基于此,借助算法自动地对神经网络进行架构的搜索成为了研究热点.神经架构搜索的方法涉及3个方面：搜索空间、搜索策略、性能评估策略.通过搜索策略在搜索空间中选择一个网络架构,借助性能评估策略对该网络架构进行评估,并将结果反馈给搜索策略指导搜索策略选择更好的网络架构,通过不断迭代得到最优的网络架构.为了更好地为读者提供一个快速了解神经网络架构搜索方法的导航地图,从搜索空间、搜索策略和性能评估策略3个方面对现有典型的神经架构搜索方法进行了梳理,总结讨论了近年来常见的架构搜索方法,并分析了各种方法的优势和不足.     

并联混合神经网络模型及应用研究

单一神经网络难以对复杂模型做出准确的预测,提出了一种并联型混合神经网络模型用于对复杂的系统进行预测,该模型由径向基函数网络、BP网络和控制模块组成。控制模块用于线性映射层,将两种单一神经网络的输出结合并得到最终的输出结果。详细地给出了混合模型的预测方法：首先,利用改进算法分别训练径向基函数网络和BP网络;其次,采用自适应遗传算法优化线性映射层以获得更好的预测精度;最后,利用两个实例比较单一神经网络和提出的混合网络的预测性能。实验表明,混合神经网络在预测精度上比单一网络具有更优的性能,同时,该混合模型为复杂系统提供了一种通用的预测工具。     

基于蒙特卡罗贝叶斯推理的移动群智感知可靠任务分配机制

针对现有任务分配策略的不足,研究了在工人数量有限的移动群智感知系统中任务分配策略,借助社交网络来分配任务并获得高收益。首先,建立了社交网络的动态不确定环境,利用社交网络完成任务,传播任务。然后考虑到不同社交网络对任务的偏好不同,设置任务偏好度这一不确定指标,借助经济学风险价值的理论描述任务分配的可靠性。最后利用蒙特卡罗贝叶斯推理方法研究任务动态传播模型的复杂参数的高斯过程,设计基于知识梯度的采样算法选择蒙特卡罗采样点,从而实现高收益的任务分配方案。为了验证所提策略的性能,将其与四种基准的采样算法进行比较。实验结果表明,所提任务分配策略在提高收益方面是有效的。     

Mixed Equilibria with Common Constraints on Transportation Networks

This study concerns about modeling the mixed equilibrium (ME) problem, including user equilibrium, system optimum, and Cournot-Nash players, with general common constraints (CCs) on transportation networks. The CCs capture the interactions of the decision variables of different players in ME, which could be internal interactions such as road link capacity constraints or external such as emission or congestion control policies. It is shown that ME with CCs can be modeled as a generalized Nash equilibrium problem (GNEP). The study proves that, under certain conditions, the GNEP-based ME is jointly convex, which can be reformulated as a variational inequality (VI). We then study the solution existence, uniqueness and solving method for the VI-based model, followed by a discussion on its potential applications. Numerical tests are conducted with common nonlinear link emission constraints as the CCs on a simple two-node, three-link network first, and then on the Nguyen Dupus network. The results show that modeling users’ route choice behavior with CCs is more general in evaluating system performance, planning link capacities, and making congestion or emission control related policies.     

物流资源整合系统的两阶段利益分配策略研究

为获取公平合理的利益分配方案,提出了基于物流资源稀缺性和任务绩效的物流资源整合系统两阶段利益分配模型。第一阶段考虑物流资源稀缺性,针对一般资源提供商和战略资源提供商分别采用一级密封价格拍卖、Shapley值法的分配策略,获取初始利益分配方案;第二阶段考虑物流资源提供商完成物流子任务的绩效对初始利益分配方案进行修正,建立物流子任务的绩效评价指标体系,利用模糊综合评价法计算物流子任务的绩效系数,获得最终的利益分配方案。以商品车配送任务为例验证该利益分配模型不仅有助于实现物流资源的优化配置,而且激励物流资源提供商提高物流任务绩效水平。     

The Relational Structure of Belief Networks

This paper demonstrates the relational structure of belief networks by establishing an extended relational data model which can be applied to both belief networks and relational applications. It is demonstrated that a Markov network can be represented as a generalized acyclic join dependency (GAJD) which is equivalent to a set of conflict-free generalized multivalued dependencies (GMVDs). A Markov network can also be characterized by an entropy function, which greatly facilitates the manipulation of GMVDs. These results are extensions of results established in relational theory. It is shown that there exists a complete set of inference rules for the GMVDs. This result is important from a probabilistic perspective. All the above results explicitly demonstrate that there is a unified model for relational database and probabilistic reasoning systems. This is not only important from a theoretical point of view in that one model has been developed for a number of domains, but also from a practical point of view in that one system can be implemented for both domains. This implemented system can take advantage of the performance enhancing techniques developed in both fields. Thereby, this paper serves as a theoretical foundation for harmonizing these two important information domains.     

计算机网络的建模与分析

计算机网络具有既随机又相关的特点，因此选择计算机网络性能评价的数学工具要考虑到网络数据流的特性，排队模型是一种确定性的数学模型，用确定性的模型描述网络既随机又相关的动态行为是不够的。Ｐｅ＃簦颍? Ｎｅｔｓ（ＰＮ）能够比较方便地刻画网络的相关录枋鐾绲木赫⑴鲎埠妥枞樱校谓肜┏浞⒄苟吹乃婊校危梢员冉戏奖愕目袒缡录乃婊裕呓岷掀鹄创酉嗟背潭壬匣汉土思扑慊缃Ｓ敕治鲋涞     

Aggregated mobility-based topology inference for fast sensor data collection

We investigate the problem of efficient data collection in wireless sensor networks where both the sensors and the sink move. We especially study the important, realistic case where the spatial distribution of sensors is non-uniform and their mobility is diverse and dynamic. The basic idea of our protocol is for the sink to benefit of the local information that sensors spread in the network as they move, in order to extract current local conditions and accordingly adjust its trajectory. Thus, sensory motion anyway present in the network serves as a low cost replacement of network information propagation. In particular, we investigate two variations of our method: a) the greedy motion of the sink towards the region of highest density each time and b) taking into account the aggregate density in wider network regions. An extensive comparative evaluation to relevant data collection methods (both randomized and optimized deterministic), demonstrates that our approach achieves significant performance gains, especially in non-uniform placements (but also in uniform ones). In fact, the greedy version of our approach is more suitable in networks where the concentration regions appear in a spatially balanced manner, while the aggregate scheme is more appropriate in networks where the concentration areas are geographically correlated. We also investigate the case of multiple sinks by suggesting appropriate distributed coordination methods.     

对等网中基于种群进化的信誉模型

信誉系统是应对对等网信任危机的关键途径。提出了一种新的基于种群进化的信誉模型。该模型将对等网视为社会生态系统,将节点的信誉评佑过程模型化为一种进化过程,通过引入交又和变异,节点能快速有效地评估潜在交易对象的信任度,从而最终提高其网络适应能力。通过仿真对模型进行了验证。     

An integrated System Dynamics model for strategic capacity planning in closed-loop recycling networks: A dynamic analysis for the paper industry

Recycling activities have demonstrated a remarkable increase over the last decade due to the economic and environmental dimensions of sustainability. In particular, capacity planning in production facilities has become a strategic issue of key importance affecting the profitability of the recycling industry.By integrating the simulation discipline and the feedback control theory into a dynamic consideration of recycling networks, this paper proposes a System Dynamics (SD) model for strategic capacity planning in the recycling industry. The decision-making process is based on a balanced tradeoff between profit and capacity utilization for a single producer with closed-loop recycling activities. The SD model captures physical stocks and flows apparent in real-world recycling networks and includes the feedback mechanisms which regulate these flows. When used as an “experimental tool”, the model tests alternative capacity planning policies and demonstrates policy suggestions for the forward and reverse channels, which maximize profitability over a strategic planning horizon. This experimentation is illustrated by using data from a paper producer with recycling activities, as a real-world test case. Extensive simulation runs, investigate the efficiency of a wide range capacity acquisition decisions, using total company profit as the measure of performance. Although such an analysis may differ from one recycling network to another, it has been kept as generic as possible to facilitate its applicability to a wide-spectrum of real-world local, regional or global networks.     

Empirical study on multiclass classification‐based network intrusion detection

Early and effective network intrusion detection is deemed to be a critical basis for cybersecurity domain. In the past decade, although a significant amount of work has focused on network intrusion detection, it is still a challenge to establish an intrusion detection system with a high detection rate and a relatively low false alarm rate. In this paper, we have performed a comprehensive empirical study on network intrusion detection as a multiclass classification task, not just to detect a suspicious connection but also to assign the correct type as well. To surpass the previous studies, we have utilized four deep learning models, namely, deep neural networks, long short‐term memory recurrent neural networks, gated recurrent unit recurrent neural networks, and deep belief networks. Our approach relies on the pretraining of the models by exploiting a particle swarm optimization–based algorithm for their hyperparameters selection. In order to investigate the performance differences, we also included two well‐known shallow learning methods, namely, decision forest and decision jungle. Furthermore, we used in our experiments four datasets, which are dedicated to intrusion detection systems to explore various environments. These datasets are KDD CUP 99, NSL‐KDD, CIDDS, and CICIDS2017. Moreover, 22 evaluation metrics are used to assess the model's performance in each of the datasets. Finally, intensive quantitative, Friedman test, and ranking methods analyses of our results are provided at the end of this paper. The results show a significant improvement in the detection of network attacks with our recommended approach.     

The performance of the Cedar multistage switching network

While multistage switching networks for vector multiprocessors have been studied extensively, detailed evaluations of their performance are rare. Indeed, analytical models, simulations with pseudosynthetic loads, studies focused on average-value parameters, and measurements of networks disconnected from the machine, all provide limited information. In this paper, instead, we present an in-depth empirical analysis of a multistage switching network in a realistic setting: We use hardware probes to examine the performance of the omega network of the Cedar shared-memory machine executing real applications. The machine is configured with 16 vector processors. The analysis suggests that the performance of multistage switching networks is limited by traffic nonuniformities. We identify two major nonuniformities that degrade Cedar's performance and are likely to slow down other networks too. The first one is the contention caused by the return messages in a vector access as they converge from the memories to one processor port. This traffic convergence penalizes vector reads and, more importantly, causes tree saturation. The second nonuniformity is the uneven contention delays induced by a relatively fair scheme to resolve message collisions. Based on our observations, we argue that intuitive optimizations for multistage switching networks may not be the most cost-effective ones. Instead, we suggest changes to increase the network bandwidth at the root of the traffic convergence tree and to delay traffic convergence up until the final stages of the network     

Paralleled hardware annealing for optimal solutions on electronicneural networks

Three basic neural network schemes have been extensively studied by researchers: the iterative networks, the backpropagation networks, and the self-organizing networks. Simulated annealing is a probabilistic hill-climbing technique that accepts, with a nonzero but gradually decreasing probability, deterioration in the cost function of the optimization problems. Hardware annealing, which combines the simulated annealing technique with continuous-time electronic neural networks by changing the voltage gain of neurons, is discussed. The initial and final voltage gains for applying hardware annealing to Hopfield data-conversion networks are presented. In hardware annealing, the voltage gain of output neurons is increased from an initial low value to a final high value in a continuous fashion which helps to achieve the optimal solution for an optimization problem in one annealing cycle. Experimental results on the transfer function and transient response of electronic neural networks achieving the global minimum are also presented.     

Improving BDD-based attractor detection for synchronous Boolean networks

Boolean networks are an important formalism for modelling biological systems and have attracted much attention in recent years. An important challenge in Boolean networks is to exhaustively find attractors, which represent steady states of a biological network. In this paper, we propose a new approach to improve the efficiency of BDD-based attractor detection. Our approach includes a monolithic algorithm for small networks, an enumerative strategy to deal with large networks, a method to accelerate attractor detection based on an analysis of the network structure, and two heuristics on ordering BDD variables. We demonstrate the performance of our approach on a number of examples and on a realistic model of apoptosis in hepatocytes. We compare it with one existing technique in the literature.     

An Efficient Handoff Decision Algorithm for Vertical Handoff Between WWAN and WLAN

Vertical handoff is one significant challenge for mobility management in heterogeneous wireless networks. Compared with horizontal handoff, vertical handoff involves different wireless network technologies varying widely in terms of bandwidth, delay, coverage area, power consumption, etc. In this paper, we analyze the signal strength model of mobile node and present a new vertical handoff decision algorithm. This algorithm can adapt to the change of mobile node＇s velocity and improve the handoff efficiency significantly. We analyze the algorithm＇s performance and the effect of different parameters on handoff triggering. In addition, we propose three performance evaluation models and verify the algorithm＇s feasibility and effectiveness in simulations.     

Service agent networks in cloud manufacturing: Modeling and evaluation based on set-pair analysis

In a cloud manufacturing system, manufacturing services form the service network and collaboration network. The performance of the networks can affect the cloud manufacturing system. Researchers can study the behavior of enterprises, service, resource and elements in cloud manufacturing through the simulation platform. In this paper, the service agent in the simulation platform can drive the manufacturing service to form a self-organization network, in order to conduct transactions and cooperation spontaneously. Thus, the service network of service agent and the collaboration network of service agent are formed. And the models of two kinds of networks are presented by the theory of Set Pair Analysis (SPA). In addition, an evaluation method for two kinds of network models is proposed. The simulation data of the networks are produced by the method of data generation in cloud manufacturing simulation platform. Finally, the evaluation methods of the networks are validated by the simulation data.