Federated learning (FL) can train a model collaboratively through multiple remote clients without sharing raw data. The challenge of federated learning (FL) is how to decrease network transmissions. This article aims to reduce network traffic by transmitting fewer neural network parameters. We first investigate similarities of different corresponding layers of convolutional neural network (CNN) models in FL, and find that there is a lot of redundant information in its model feature extractors. For this, we propose a communication-efficient federated aggregation algorithm named FedSL (Federated Split Layers) to reduce the communication overhead. Based on the number of global model layers, the FedSL divides client models into groups in the depth dimension. A Max-Min client selection strategy is employed to select participants for each layer. Each client only transfers partial parameters of those layers that are selected, which reduces the number of parameters. FedSL aggregates the global model in each group and concatenates the parameters of all groups according to the order of layers. The experimental results demonstrate that FedSL improves communication efficiency compared to the algorithms (e.g., FedAvg, FedProx, and MOON), decreasing 42% communication cost with VGG-style CNN and 70% with ResNet-9, while maintaining a similar model accuracy with baseline algorithms.

The governance technology of decentralized autonomous organizations (DAOs) provides an effective solution for solving existing organizational management issues. Governance mechanisms of DAOs are usually encoded in smart contracts in the form of rule sets and executed automatically. However, the mechanism’s logical flaws and code errors expose DAOs to unpredictable risks. Complex adaptive dynamics computer-aided design (CadCAD) can test the effectiveness of the mechanisms through simulation. Nonetheless, as DAOs are typical complex systems with social and engineering complexity, managing, controlling, and supervising their operation through traditional methods are difficult. The parallel intelligence theory based on artificial societies, computational experiments, and parallel execution (ACP) method provides an effective research framework and practical method for solving DAOs’ governance issues. Therefore, in this article, we propose a parallel mechanism verification method and execution system, namely, complex adaptive systems for computer-aided dynamic design (casCAD2) as an extension of cadCAD. Leveraging parallel intelligence and cyber–physical–social systems (CPSS), casCAD2 is capable of probing into the laws that govern system evolution within a simulated environment. It serves as a robust tool for verifying the efficacy of DAOs’ mechanisms and predicting their potential risks. We also build a parallel market-based anchoring mechanism (MAM) system to demonstrate how it can be used for DAOs’ mechanism verification. This study can provide a new research method and application system for DAOs’ effective governance.

Recent years have witnessed remarkable developments and increasingly deepened integrations between blockchain as a decentralized computing architecture and auction as an efficient resource allocation approach. Typically, blockchain can help provide a secured and trusted distributed environment for various auction scenarios, while auction is particularly suitable for designing resource allocation and pricing mechanisms in blockchain systems. As such, integrative research on blockchain and auction developed rapidly and attracted widespread attention in various fields ranging from academia to financial, industrial, and social services. However, a comprehensive survey on this interdisciplinary topic is still nonexistent, which motivates our work. In this article, we aim to fill this important research gap by reviewing the related literature. We first conducted a brief overview of blockchain technology and auction theory, and then systematically discussed the research progress on the existing blockchain research based on auction theory as well as auction research enabled by blockchain. Toward the end, we presented several open research issues and directions, aiming to provide useful guidance and reference for future research efforts.

With the rapid development of blockchain technology and industries, scalability has been widely realized as one of the primary and urgent concerns for the large-scale adoption of blockchain, especially for cryptocurrencies. In this respect, directed acyclic graph (DAG) proves to be an elegant solution to scaling blockchain but suffers from weak consistency and security issues. In this article, we designed a novel DAG-BLOCK architecture for blockchain-enabled cryptocurrency markets in order to improve the scalability. In our work, DAG is used to replace the Merkel-tree-based transaction structure within the block, and a novel design of open blocks is proposed to enable user nodes to participate in verifying the transactions in blockchain systems. On this basis, we designed a new segmented market structure, in which each miner serves only a group of users instead of all users, so as to reduce miners’ workload and thus scale transaction processing capabilities. Our work can help improve the scalability of cryptocurrencies via evolving the underlying blockchain systems to graph-based distributed ledgers and is expected to shed new light on designing blockchain-based decentralized markets.

To break the dilemma of data island, a distributed framework for trustable control is proposed toward information security and data privacy in large-scale cyber-physical systems. The federated control system consists of distinct blockchains, as such information security and data privacy are technologically guaranteed. Moreover, data are divided into private and nonprivate data. Only nonprivate data can be exchanged for a better global system performance, where the interblockchain communication is ensured by cross-blockchain technologies. Federated control framework establishes a trustable environment where each subsystem is willing to share data for optimal performance. The architecture, structure, and implementation process of federated control are discussed, together with the potential applications to smart buildings.

Current evidence indicates that the semantic representation of question and answer sentences is better generated by deep neural network-based sentence models than traditional methods in community answer selection tasks. In particular, as a widely recognized language model, the self-attention model computes the similarity between the specific word and the whole sets of words in the same sentence and generates new semantic representation through the similarity-weighted summation of semantic representations of the whole words. However, the self-attention operation entirely considers all the signals with a weighted sum operation, which disperses the distribution of attention, which may result in overlooking the relation of neighboring signals. This issue becomes serious when applying the self-attention model to online community question answering platforms because of the varied length of the user-generated questions and answers. To address this problem, we introduce an attention mechanism enhanced local self-attention (LSA), which restricts the range of original self-attention by a local window mechanism, thereby scaling linearly when increasing the sequence length. Furthermore, we propose stacking multiple LSA layers to model the relationship of multiscale$n$-gram features. It captures the word-to-word relationship in the first layer and then captures the chunk-to-chunk (such as lexical$n$-gram phrases) relationship in its deeper layers. We also test the effectiveness of the proposed model by applying the learned representation through the LSA model to a Siamese and a classification network in community question answer selection tasks. Experiments on the public datasets show that the proposed LSA achieves a good performance.

Crypto management is proposed to tackle the management decision-making challenges under data asymmetry and trust asymmetry that cannot be solved merely by technical means. It emphasizes the novel management model for the real-time generation of reliable, trustworthy, and usable management decisions based on blockchain and blockchain-driven technologies. First, the framework model of crypto management with detailed descriptions of each technique is introduced, where blockchain is the underlying technology, decentralized autonomous organization (DAO) is the management structure, federated data is the decision basis, smart contract is the decision method, and non-fungible token (NFT) is the main decision incentive. Then, its collaboration mechanisms of on-blockchain DAO and off-blockchain organization as well as intra-organization and extra-organization nodes are discussed. Moreover, the potential applications of crypto management are addressed, and a case of task-oriented performance management is given to state how crypto management works to generate the real-time management decisions. Toward the end, the future research directions are pointed out in this emerging new area.

Artificial intelligence (AI) has been witnessed to provide valuable solutions to all walks of life. However, data island and computing resources limitations in the centralized AI architectures have increased their technical barriers, and thus distributed AI collaboration in data, models, and resources has attracted intensive research interests. Since the existing trust-based collaboration models are no longer applicable for the large-scale distributed collaboration among trustless machines in open and dynamic environments, this article proposes a novel decentralized AI collaboration framework, i.e., learning markets (LM), in which blockchain provides a trustless environment for collaboration and transaction, while smart contracts serve as software-defined agents to encapsulate and process scalable collaboration relationships and market mechanisms. LM can not only help those participants without mutual trust realize collaborative mining with dynamic and quantitative rewards but also build an AI market with natural auditability and traceability for trading trusted and verified models. We implement and comprehensively analyze LM based on the Ethereum interplenary file system platform (IPFS), and the results prove that it has advantages in collaboration fairness, transparency, security, decentralization and universality. Based on our collaboration framework, distributed AI contributors are expected to cooperate and complete those learning tasks that cannot be done previously due to lack of complete data, sufficient computing resources and state-of-the-art models.

The emerging applications for smart cities intend to promote the quality of citizens’ life. Among them, ubiquitous user connectivity and real-time computation offloading are significant for the ever-increasing requirements of delay-sensitive and mission-critical applications. By integrating human social behaviors (such as relationship, similarity, community, and social ties) with physical Internet of Things (IoT) systems, social IoT systems are promising to provide ubiquitous connectivity among users. As the applications of social IoT systems are transferring from information dissemination to user entertainment (such as image identification, online games, and augmented reality), computation offloading is significant to reduce the execution delay of applications.

Bitcoin is gaining ground in recent years. In the Bitcoin system, miners provide computing power to confirm transactions in pursuit of transaction fees, while users compete by bidding transaction fees for faster confirmation. This process is in essence analogous to online ad auctions, where advertisers bid for more prominent ad slots. Therefore, inspired by ad auction research, we propose to apply the Generalized Second Price (GSP) auction mechanism in the dynamic confirmation game on Bitcoin transactions. Our model is targeted to deal with the problems caused by instability and low efficiency in the currently-adopted Generalized First Price (GFP) auction model in Bitcoin confirmation games. Besides, we use the “rank-by-cost” rule to replace the “rank-by-fee” rule, where each transaction’s cost is calculated by the user-submitted fee and the waiting time. Aiming to probe users’ equilibrium strategy, we first discuss the GSP game with complete information under synchronous submissions, and show that it has the Locally Envy-Free equilibrium. Then, we study the GSP game with incomplete information under asynchronous submissions, and define two types of strategies, i.e., the Farsighted Balanced (FB) strategy and the Instant Balanced (IB) strategy. The FB strategy is in line with users’ practical needs of determining fees so as to maximize the long-term payoffs; however it cannot generate a stable equilibrium. Alternatively, the IB strategy focuses on the instant payoff maximization, and if all users follow the IB strategy, their equilibrium fees can finally converge to a stable profile. Finally, we design computational experiments to validate our theoretical models and analysis. Our research findings indicate that this novel GSP mechanism is superior to the currently adopted GFP mechanism. Besides, the convergence of the GSP game under the IB strategy has also been illustrated by the computational experiments.

Urban transportation systems are shaped by factors that include people, vehicles, roads, and the environment, forming a complex and giant system with dynamics, diversity, and uncertainty. Physical signal-driven intelligent transportation systems (ITSs) typically lack the ability to capture social behaviors or crowd willingness, and they achieve only information automation for transportation decision support. The crowdsourcing social signals consist of timely, extensive, comprehensive, and rich intelligence that concern urban dynamics, social behaviors, and traffic environments. Such social signals provide a new paradigm for operating ITS with unstructured semantic data, making knowledge automation for decision intelligence a possibility. This article reviews the knowledge automation paradigms for cyber-physical-social systems (CPSSs) compared with traditional information automation paradigms for cyber-physical systems (CPSs) in ITS, from the perspective of data-driven, modeling space, analytical methodologies, and decision support services. To investigate the key methodology in social spaces that enhance information automation into knowledge automation, we summarize the current research into a multisource heterogeneous social signal-based traffic decision knowledge automation framework and further exploit the computational paradigm and applications scenarios of this framework. Finally, we discuss future challenges for designing and realizing knowledge automation on CPSS in transportation.

Welcome to the fifth issue of Ieee Transactions on Computational Social Systems (TCSS) of 2021. As usual, we would like to share some great news first. Since April 2021, IEEE TCSS has been added to the Science Citation Index Expanded (SCIE) database in Clarivate Web of Science. We are excited to report that all TCSS articles published since 2018 have been backtracked and indexed by SCIE.

Since early 2020, the COVID-19 global pandemic has significantly impacted almost every aspect of the human society throughout the world. Until now, middle of 2021, although with all the efforts on pandemic intervention and vaccination, COVID-19 is still hovering around the world, resulting in more than 177 million confirmed cases and 3.8 million deaths.

In proof-of-work (PoW) consensus protocol-based blockchain mining, the pools can increase their rewards by utilizing block withholding attack. As such, how much computational power should be used to attack other pools becomes an important decision issue faced by the pools. This article mainly studies the block withholding attack issue faced by mining pools. Considering the case that there are two pools, where only one pool can attack the other pool, we propose an optimal block withholding strategies for pools. We also illustrate that attacking is not always the optimal strategies for the pools and present the conditions for attacking. With computational experiment approach, we designed several experiments to validate our proposed strategies, and our results can provide important managerial insights for pools in blockchain mining.

Traditional organization management typically follows a top-down pyramid structure, which is widely believed to have many problems in releasing innovation potentials. In the new era of intelligent technologies, knowledge automation is required to meet the urgent demand for rapid acquisition and application of knowledge. With the rapidly deepened integration of the real world and the virtual society, cyber-physical-social system (CPSS)-oriented parallel management proves to be an effective and efficient way in solving these problems. In this article, we utilize blockchain technology and smart contracts in knowledge automation and investigate blockchain-based knowledge automation, which can be used for CPSS-oriented parallel management. We also propose a management framework based on the smart contract and discuss a case study.

Welcome to the first issue of the IEEE Transactions on Computational Social Systems (TCSS) of 2020, and Happy New Year to You! We would like to take this opportunity to express our sincere thanks to our editors, reviewers, authors, and readers for your great support and effort devoted to the TCSS, along with our best wish and hope that everyone has a happy, healthy, and fruitful 2020.

Blockchain, as an emerging decentralized architecture and distributed computing paradigm underlying Bitcoin and other cryptocurrencies, has attracted intensive attention in both research and applications recently. Blockchain, especially powered by chain-coded smart contracts, has the full potential of revolutionizing increasingly centralized cyber-physical-social systems (CPSSs) for constructions and applications, and reshaping traditional knowledge automation workflows. The key advantage of blockchain technology lies in the fact that it can enable the establishment of secured, trusted, and decentralized autonomous ecosystems for various scenarios, especially for better usage of the legacy devices, infrastructure, and resources.

Proof of work and proof of stake (PoS) are commonly used in the current permissionless blockchains. These consensus protocols can be abstracted into a random process of selecting a node for accounting in a blockchain ledger. However, they are generally faced with resource consumption and vulnerability issues. We present proof of credit (PoC), a fair blockchain protocol based on the PoC blockchain protocol. It is a special PoS protocol where the credit is a special kind of stake quantifying whether the node's activity is beneficial to the system. Any nodes cannot change their credits arbitrarily. We demonstrate that our PoC protocol satisfies the security properties, including common prefix, chain quality, and chain growth, under the assumption that the total credit the honest held is majority. In addition, we propose a self-audit mechanism and a hybrid incentive mechanism to enhance the security and stability. Finally, we explain the method by which the PoC protocol resists the double-spending attacks and the selfish mining attacks.

The Internet of Things (IoT) is expected to connect a massive number of smart devices to the Internet. The existing centralized architecture for handling the huge volume of data created in the IoT is facing many research challenges, including security and privacy, trustworthiness, operational challenges, business models and the practical aspects, and legal and compliance issues. These challenges ask for new approaches to online identity, trustworthy transactions, and resilient networks.

Decentralized autonomy is a long-standing research topic in information sciences and social sciences. The self-organization phenomenon in natural ecosystems, the Cyber Movement Organizations (CMOs) on the Internet, and the Distributed Artificial Intelligence (DAI), and so on, can all be regarded as its early manifestations. In recent years, the rapid development of blockchain technology has spawned the emergence of the so-called Decentralized Autonomous Organization [DAO, sometimes labeled as Decentralized Autonomous Corporation (DAC)], which is a new organization form that the management and operational rules are typically encoded on blockchain in the form of smart contracts, and can autonomously operate without centralized control or third-party intervention. DAO is expected to overturn the traditional hierarchical management model and significantly reduce organizations’ costs on communication, management, and collaboration. However, DAO still faces many challenges, such as security and privacy issue, unclear legal status, and so on. In this article, we strive to present a systematic introduction of DAO, including its concept and characteristics, research framework, typical implementations, challenges, and future trends. Especially, a novel reference model for DAO which employs a five-layer architecture is proposed. This article is aimed at providing helpful guidance and reference for future research efforts.

Welcome to the first issue of the IEEE Transactions on Computational Social Systems (TCSS) this year, and Happy New Year to everyone. We would like to take this opportunity to express sincere gratitude to our editors, reviewers, authors, and readers for your support and great efforts devoted to TCSS. Also, we want to express the best wishes to you all, and hope you have a happy, healthy, and fruitful 2019.

Welcome to the third issue of the IEEE Transactions on Computational Social Systems (TCSS) in 2019. Thanks to the efforts of the editors, reviewers, authors, and readers of TCSS, the influence of TCSS is rapidly increasing. According to the latest statistics released by Elsevier, the CiteScore of TCSS in 2018 reaches 4.00, and ranks eighth out of the 255 journals (top 3%) in the field of social sciences. This is a solid improvement compared with the corresponding data in 2017 (CiteScore: 2.36, Rank: 17/226, and top 8%). Thanks and congratulations to our authors, reviewers, and members of our editorial board. The current issue includes 20 regular papers and a brief discussion on social energy.

In recent years, the rapid development of cryptocurrencies and their underlying blockchain technology has revived Szabo’s original idea of smart contracts, i.e., computer protocols that are designed to automatically facilitate, verify, and enforce the negotiation and implementation of digital contracts without central authorities. Smart contracts can find a wide spectrum of potential application scenarios in the digital economy and intelligent industries, including financial services, management, healthcare, and Internet of Things, among others, and also have been integrated into the mainstream blockchain-based development platforms, such as Ethereum and Hyperledger. However, smart contracts are still far from mature, and major technical challenges such as security and privacy issues are still awaiting further research efforts. For instance, the most notorious case might be “The DAO Attack” in June 2016, which led to more than $50 million Ether transferred into an adversary’s account. In this paper, we strive to present a systematic and comprehensive overview of blockchain-enabled smart contracts, aiming at stimulating further research toward this emerging research area. We first introduced the operating mechanism and mainstream platforms of blockchain-enabled smart contracts, and proposed a research framework for smart contracts based on a novel six-layer architecture. Second, both the technical and legal challenges, as well as the recent research progresses, are listed. Third, we presented several typical application scenarios. Toward the end, we discussed the future development trends of smart contracts. This paper is aimed at providing helpful guidance and reference for future research efforts.

Bitcoin is a novel protocol with the potential of enabling a decentralized and trustless cryptographic currency, and its underlying technology named blockchain operates on a worldwide basis via a complex set of rules originally proposed by Nakomoto in 2008. In Bitcoin blockchain, miners provide computational services (i.e. mining) to get profits from the fixed rewards of newly found block and also transaction fees from recording the users’ transactions to the blocks. With the decreasing of the fixed new block reward, transaction fees will play the role as the main profit source of miners, thus provide important supports for the sustainability and vitality of the Bitcoin system. Therefore, it is of great necessity to research transaction fees. In this paper, we investigate transaction fees in a queuing game with non-preemptive priority, in which both the miners’ mining rewards and the users’ time cost are highlighted. Then, we conduct theoretical analysis of the game, getting five types of Nash equilibria of the game. We also find that the over-long waiting time will bring negative marginal profits on transaction fees to some users with low time cost, therefore, they will not be willing to offer transaction fees.

With the development of the blockchain technology, Bitcoin mining has become more and more popular. This paper aims to provide a three-level framework of the economic issues in Bitcoin mining research, from the levels of mining pools, individual miners and blockchain network. We also offer an overview of relevant research efforts in literature. Considering the uncertainty, diversity and complexity of the Bitcoin ecosystems, we propose a novel research framework based on the ACP theory, which can be used to explore the economic issues in the Bitcoin ecosystems. This paper aims to provide a preliminary investigation to the economic issues faced by participants in the Bitcoin ecosystems, and stimulate the attentions and interests of researchers in this field.

With the rapid development of cryptocurrency and its underlying blockchain technologies, platforms such as Ethereum and Hyperledger began to support various types of smart contracts. Smart contracts are computer protocols intended to digitally facilitate, verify, or enforce the negotiation or performance of a contract. Smart contracts have broad range of applications, such as financial services, prediction markets and Internet of Things (IoT), etc. However, there are still many challenges such as security issues and privacy disclosure that await future research. In this paper, we present a comprehensive overview on blockchain powered smart contracts. First, we give a systematic introduction for smart contracts, including the basic framework, operating mechanisms, platforms and programming languages. Second, application scenarios and existing challenges are discussed. Finally, we describe the recent advances of smart contract and present its future development trends, e.g., parallel blockchain. This paper is aimed at providing helpful guidance and reference for future research efforts.

In recent years, social media platforms and instant messaging applications have been witnessed to play an increasingly important role as a communication channel among employees to build effective organizations, which results in a considerably large part of individual activities moving onto the cyber-based workspace. The online behaviors of employees rely heavily on the policies or mechanisms formulated for the cyber-space, and are expected to impose a significant influence on their performance appraisal in the physical world. In this paper, we focus on the typical online reporting behavior, and empirically study how the work reporting policies formulated by supervisors influence the employees' strategies and in turn their performance, using a unique real-world dataset collected from the Chinese social media platform called WeChat. We consider two types of work reporting policies: the one is concerned with the reporting time and the other with the duplication degree of reporting contents. First, we establish an optimization model, taking both the reputation-based and performance-related utilities into consideration, so as to examine the employees' reporting strategies and their actual work performance. On this basis, we make further discussions about the reporting policy optimization from the perspective of organization supervisors. Then, an empirical study on a median-size organization in China is conducted to validate our proposed model and analysis. The results prove our conclusion that the reporting policy will influence not only employees' reporting strategies, but also their actual work performance.

With the increasing difficulty of solo mining in blockchain mining, pool mining has become more and more popular, and most of the miners would like to join a mining pool and contribute their computational power to the pool. When the pool finds a valid block and get the reward from the blockchain network, it will distribute the reward to its miners according to its reward mechanism. In practice, the Pay-Per-Last-N-Shares (PPLNS) mechanism is one of the most commonly used mechanisms by pools, and the pool adopting PPLNS mechanism will distribute the reward to the miners whose reported shares are in the last N shares, according to their proportion of the number of shares in the last N shares. In the PPLNS mechanism, different reporting strategies may bring different rewards for miners. Thus, how to report their found shares to the pool has become an important issue faced by the miners. In this paper, we study the share reporting problem faced by the miners in PPLNS pools, and establish a share reporting optimization model for the miners. We also study the effect of the parameter N in the PPLNS mechanism on the optimal reporting strategies of the miners. With the computational experiments approach, we design experiments to evaluate our proposed share reporting strategies. This work is the first attempt to study the share reporting issue faced by miners in PPLNS pools, and it can provide useful managerial insights for miners when making their share reporting decisions in such pools.

In recent years, social media platforms and, especially, instant messaging applications have been witnessed to play an increasingly important role as a communication channel among employees to build effective organizations, which results in a considerably large part of individual activities moving onto the cyber-based workspace. In this paper, we collect the unique real-world data set from a WeChat-based work performance evaluation system built and used by a medium-sized organization in China, in order to study the optimization of the employees’ work strategies, especially the work time determination, with the purpose of maximizing the total utility comprised of the work and nonwork utilities. In the system, the performance evaluation is completely based on the employees’ work reports and includes three major measures including score, level, and rank. First, we formulate a dynamic optimization model, incorporating the work utility into the reputational and the substantive utility, so as to examine the employees’ work strategy. Then, we conduct both empirical studies and computational experiments to make an in-depth analysis to probe the employees’ optimal work strategies. The main results include: 1) the optimal work time is greatly affected by the weight of the reputational utility; 2) the optimal work time is always the threshold time to achieve a certain level or a certain rank; 3) the total utility is with the trend of decreasing after increasing with the growing work time; and 4) the optimal work time does not always make the employee achieve a good work performance.

With the application of big data analytics in online marketing, real-time bidding (RTB) has developed to be the primary business model and also the major online advertising channel. Due to the precise analysis of Web Cookies, RTB platforms can target the visiting audiences and then forward their generated ad impressions to demanding advertisers who bid on the best-matched audience in a real-time fashion. In RTB markets, the reserve price plays the vital role as a tuner to exclude over-low bids, and hence guarantee the desirable sales prices and revenues for publishers from ad impression sales. In this paper, we strive to study the publisher's strategy on the reserve price and probe its impact on his/her revenues. We first discuss the reserve price of ad impression in a single-channel sales model, including the online RTB channel or the off-line direct channel, aimed to study its impact on the publisher' revenue. Then, we further analyze the impact of the reserve price in the multi-channel settings. Finally, we conduct experiments using empirical log data collected from real-world RTB markets to validate our models and analyses, and the experimental results indicate that: 1) in the single-channel sales model, publishers should set the reserve price for only the online-channel ad impressions while not for the off-line-channel ones and 2) in the multi-channel ad impression sales, publishers should set both off-line and online reserve prices for revenue maximization.

Cyber-physical social system (CPSS) plays an important role in both the modern lifestyle and business models, which significantly changes the way we interact with the physical world. The increasing influence of cyber systems and social networks is also a high risk for security threats. The objective of this paper is to investigate associated risks in CPSS, and a hybrid Bayesian risk graph (HBRG) model is proposed to analyze the temporal attack activity patterns in dynamic cyberphysical social networks. In the proposed approach, a hidden Markov model is introduced to model the dynamic influence of activities, which then be mapped into a Bayesian risks graph (BRG) model that can evaluate the risk propagation in a layered risk architecture. Our numerical studies demonstrate that the framework can model and evaluate risks of user activity patterns that expose to CPSSs.

Since the increasing popularization of the emerging blockchain technology, blockchain mining has attracted more and more attention. Due to the difficulty of solo mining, typically miners choose to join a mining pool. As there are many mining pools and different mining pools may adopt different reward mechanisms, how to choose the appropriate mining pool has become one of the most important issues faced by miners, since miners can get different rewards in different pools. In practice, there are three commonly used reward mechanisms for the mining pools to distribute the reward among their miners, namely, the proportional mechanism, the pay-per-share mechanism, and the pay-per-last-N-share mechanism. In this paper, we study the pool selection problem faced by the miners, and model it as a risk decision problem since different reward mechanisms have different risks. We establish a pool selection model based on the maximum-likelihood criterion and also study the effect of N on the miners' optimal pool selection decisions. By utilizing the computational experiments approach, we validate our proposed pool selection strategies. Our results can provide important managerial insights for miners when making their pool selection decisions.

To improve the accuracy of diagnosis and the effectiveness of treatment, a framework of parallel healthcare systems (PHSs) based on the artificial systems + computational experiments + parallel execution (ACP) approach is proposed in this paper. PHS uses artificial healthcare systems to model and represent patients’ conditions, diagnosis, and treatment process, then applies computational experiments to analyze and evaluate various therapeutic regimens, and implements parallel execution for decision-making support and real-time optimization in both actual and artificial healthcare processes. In addition, we combine the emerging blockchain technology with PHS, via constructing a consortium blockchain linking patients, hospitals, health bureaus, and healthcare communities for comprehensive healthcare data sharing, medical records review, and care auditability. Finally, a prototype named parallel gout diagnosis and treatment system is built and deployed to verify and demonstrate the effectiveness and efficiency of the blockchain-powered PHS framework.

Crime scene analysis is a retrospective process from traces to psychology and physiology. It is not only the starting point and foundation of criminal investigation, but also the key part for solving criminal cases. As a typical open complex social system, it has three features, namely, uncertainty, diversity, and complexity, thus making the system modeling a huge challenge. In this paper, we propose the parallel crime scene analysis system based on the artificial societies, computational experiments and parallel execution (ACP) approach, which uses artificial (A) crime scene to describe the basic elements, functions and states of the criminals, computational (C) experiments to compute and predict the different forms of crime scene, and parallel (P) execution to guide or control the evolution of the physical crime process in accordance with the results from the artificial crime scene. First, we propose the concept of parallel crime scene from the perspective of complex system theory and give an overview of its architecture, then we present the construction method of artificial crime scene and the blackboard-based multiagent artificial crime scene analysis system. On this basis, the temporal and spatial interaction models of the criminal subjects are proposed and verified. After that, we introduce the software-defined crime scene analyzing model systematically. The ACP approach sheds light on the intelligent management and control for complex crime scene analysis.

Welcome to the third issue of the IEEE Transactions on Computational Social Systems (TCSS) for 2018.

Time flies fast, it has been already one year since I was appointed as the Editor-in-Chief of this great publication, and thanks to the strong support and dedication of our associate editors, editorial staff, anonymous reviewers, and authors, we have made solid progress and I really enjoy my work and our achievement so far. At this point, significant improvements in the timeliness and quality of the review process, as well as the numbers of manuscripts submitted and articles published have been accomplished.

Welcome to the first issue of the IEEE Transactions on Computational Social Systems (TCSS) for 2018, and Happy New Year to everyone. According to the Chinese lunar calendar, this is the year of the Dog, which in Chinese culture represents trust, loyalty, dedication, and energy. As such, I would like to take this opportunity to express my best wishes of a happy, healthy, and high-producing 2018 to each and every one of our readers, reviewers, and editors.

Welcome to the last issue of the IEEE Transactions on Computational Social Systems (IEEE TCSS) in 2018. Starting from the first issue next year, our Transactions will be a bimonthly publication, entering a new stage for the IEEE TCSS.

This paper is to discuss the state, trend, and frontiers of development of cyber-physical-social systems (CPSSs) in China. The demand for developing CPSS is discussed in detail, followed by the Artificial societies, Computational experiments, Parallel execution (ACP) approach for CPSS and knowledge automation. The development of ACP based on CPSS in transportation, energy, information, Internet of Things, and Internet of Minds (IoM) is discussed to demonstrate the cutting-edge applications in CPSS. Finally, the blockchainized IoM technology and the concepts of parallel society are described. This paper will contribute to the transition from the current social construct to a futuristic intelligent society.

As an emerging decentralized architecture and distributed computing paradigm underlying Bitcoin and other cryptocurrencies, blockchain has attracted intensive attention in both research and applications in recent years. The key advantage of this technology lies in the fact that it enables the establishment of secured, trusted, and decentralized autonomous ecosystems for various scenarios, especially for better usage of the legacy devices, infrastructure, and resources. In this paper, we presented a systematic investigation of blockchain and cryptocurrencies. Related fundamental rationales, technical advantages, existing and potential ecosystems of Bitcoin and other cryptocurrencies are discussed, and a six-layer reference model of the blockchain framework is proposed with detailed description for each of its six layers. Potential applications of blockchain and cryptocurrencies are also addressed. Our aim here is to provide guidance and reference for future research along this promising and important direction.

With the rapid development of big data analytics in online marketing, real-time bidding (RTB) has emerged as a promising business model in recent years, and now becomes one of the major online advertising channels. Based on analysis of Web Cookies, RTB platforms are able to precisely identify the features and preferences of target audiences visiting publishers' websites, and forward the generated ad impressions to competing advertisers who submit bids for their best-matched audience in real-time ad auctions. In RTB markets, reserve price serves as an important tuner to exclude advertisers with low estimated values, and hence can guarantee a desirable result for the publisher from ad impression auctions. In this paper, we strive to study publishers' strategy on the reserve price, and probe the impact of reserve price on their revenues. We first analyze the ad impression auction under a direct auction mechanism. We then introduce the reserve price and study its impact on publishers' revenues under an indirect auction mechanism, and our research findings indicate that a rational positive reserve price will always improve publishers' revenues even if it is not optimal. Also, the optimal reserve price is figured out based on the advertisers' bid distributions for publishers' revenue maximization. Finally, experiments using empirical log data from real-world RTB markets are designed to validate our model and analysis, and the results provide strong support to our theoretical analysis. The experimental results also indicate that although the number of bids does not impose any influence on the optimal reserve price, it has significant impacts on publishers' revenues.

Real time bidding (RTB) has become an emerging online advertising with the development of Internet big data in recent years. In the whole RTB ecosystem, the Demand Side Platform (DSP) plays a central role, and it realizes the programmatic and accurate buying of the advertisements for the advertisers via a two-stage auction. In RTB business logics, DSP plays as an intermediary between the advertisers and the center platform. Due to the principle-agent relationship between the advertisers and the DSP, the DSP aims not only to maximize the revenue for the advertisers, but also gain its revenue in this process. So far there are two revenue modes for DSP, namely the two-stage resale model and the commission model, respectively. In this paper, we mainly consider the revenue model for DSP in RTB advertising market. We aim to study the properties of the two revenue models, and compare the revenues for the DSP and the advertisers under these two models. We also provide an example to illustrate our proposed models and their properties. The results show that under small ratio of the commission, the advertisers are more likely to choose the commission model, but the DSP is more likely to choose the two-stage resale model and set a larger weight, while under large ratio of the commission, the advertisers are more likely to choose the two-stage resale model, but the DSP is more likely to choose the commission model. Our research work highlights the importance of the revenue model on the revenues of the advertisers and the DSP, and is intended to provide a useful reference for DSPs in RTB advertising markets.

With the ever-growing popularization of Real Time Bidding (RTB) advertising, the Ad Exchange (AdX) platform has long enjoyed a dominant position in the RTB ecosystem due to its unique role in bridging publishers and advertisers in the supply and demand sides, respectively. A novel technology called header bidding emerged in the recent one or two years, however, is widely believed to have the potential of challenging this dominant position. Compared with RTB markets, header bidding establishes a priority sub-market allowing bidding partners of the publisher submit their bids before the ad impression delivered to the open AdX platform, resulting in a decreased winning probability and revenue for the AdX. As such, there is a critical need for the AdX to tackle this challenge so as to better coexist with header bidding platforms. This need motivates our research. We utilize stochastic programming approach and establish a stochastic optimization model with risk constraints to optimize the pricing strategy for the AdX, considering that the highest bids from the bidding partners can be characterized by random variables. We study the equivalent forms of our proposed model in case when the randomness is characterized by uniform or normal random variables. With the computational experiment approach, we validate our proposed model, and the experimental results indicate that both the risk tolerance of the AdX and the distribution of randomness of the highest bid from the bidding partners can greatly affect the optimal strategy and the corresponding optimal revenue of the AdX. Our work highlights the importance of the risk level of the AdX and the distribution of the randomness generated by the partners to the decision making process of the AdXs in header bidding markets.

In the paper, considering the limitation of effective method in E-learning area, a recommendation framework for E-Learning based on deep learning is proposed. Our model is based on deep learning, which has strong capability to learn from large-scale data. It has some improvements than previous methods. First, it is based on the conventional K-Nearnest Neighbor(KNN) method to train a model, thus its accuracy is guaranteed. Second, it can recommend the new item whose similarity can not be calculated. Third, it greatly reduces the heavy burden for a running system, which is useful in real practice of recommendation systems. In conclusion, the proposed framework can offer a new recommendation method for more personlized learning in the future.

In social marketing practice, it is usually important to anticipate the long-term impact of the target application to maintain a long-lasting marketing effect, whereas a new product or technology should spread as quickly as possible to establish a competitive advantage. To find a balance between them, we tackle this challenge by modelling the problem as an issue of time-discounted influential sustainability. Given a threshold μ, the goal of the problem is finding a small subset of nodes as seeds and deciding the optimal timing to activate each seed that could maximize the time-discounted number of iterations, each of which actives more than μ nodes. We prove that solving the problem is NP-hard and the objective function is non-negative, non-monotonic, and non-submodular. Therefore we propose a greedy approach to approximately solve this problem. Our experimental results demonstrate that our solution outperforms two baseline algorithms. In order to provide meaningful advices for advertisers on selecting proper initial seed users, we further analyze and compare the performance of four seeding strategies on three typical types of social networks.

Online advertisers bidding in sponsored search auctions through Web search engines are experiencing unprecedentedly fierce competition in recent years, resulting in an obvious upward trend in bids submitted by advertisers. This bid inflation phenomenon poses significant threat to the overall stability and the effectiveness of sponsored search markets. Existing research efforts, however, do not yet provide directly relevant theoretical or managerial insights to help understand this kind of real-world competitive bidding behavior in sponsored search. Our research is targeted at filling in this important research gap. Based on a model of advertisers' rational competitive preference, we propose a novel equilibrium solution concept called the upper bound Nash equilibrium (UBNE), which can be viewed as the upper bound of the output-truthful subset in the NE continuum of sponsored search auctions. We show that the UBNE can better characterize advertisers' competitive bidding behavior than other solution concepts, offering a viable theory driven behavioral explanation for bid inflation. We also show that the UBNE is a stable outcome of competitions among advertisers in repeated game settings and yields the optimal outcome for Web search engines.

With the rapid development of big data analytics in online marketing, real-time bidding (RTB) has emerged as a promising business model in recent years and now becomes one of the major online advertising channels. Based on analysis of Web Cookies, RTB platforms are able to precisely identify the features and preferences of target audiences visiting publishers' websites, and forward the information to competing advertisers submitting bids for their best-matched audience in real-time ad auctions. As the supplier of ad impressions, publishers typically have multiple channels to sell their ad impressions (i.e., ad inventory), making their strategies for allocating ad inventory one of the most critical research problems. In this paper, we strive to study publishers' optimal strategy of allocating ad inventory across online channel of RTB-based auctions and offline channel prevailingly realized in the form of guaranteed contracts. Considering the ad reserve price as the control variable, we establish the optimization model. We also explicitly take the default penalty in offline channels into consideration, so as to balance the short-term online revenue and long-term offline revenue. In our work, we analyze altogether three kinds of strategies for publishers to allocate their ad inventory in pursuit of the optimal strategy, and validate our model and analysis via computational experiments. We find that there is no dominant strategy that can outperform others in all cases, and interestingly, publishers using the hybrid-channel strategy do not always gain more revenues than those using the single-channel strategy.

Real Time Bidding (RTB) is an emerging business model and a popular research topic of online advertising markets. Using cookie-based big-data analysis, RTB advertising platforms have the ability to precisely identify the features and preferences of online users, segment them into various kinds of niche markets, and thus achieve the precision marketing via delivering advertisements to the best-matched users. The segmentation granularity used by such platforms, typically referred to as the Demand Side Platforms (DSPs), plays a central role in the effectiveness and efficiency of the RTB ecosystem. In practice, fine-grained user segmentations may lead to increased value-per-clicks and bid prices from advertisers, but at the same time reduced competition and possibly decreased bid prices in each niche market. This motivates our research on the optimal segmentation granularity to solve this dilemma faced by DSPs. Using a RTB market model with two-stage resales, we analyzed DSPs' segmentation strategies taking the revenues of both advertisers and DSPs into consideration. We also validated our proposed model and analysis using the computational experiment approach, and the experimental results indicate that with the increasing of segmentation granularity, the weighted sum of the DSP and advertisers' revenues tends to first rise and then decline in all weight-value cases, and the optimal granularity is greatly influenced by the value of weights. Our work highlights the need for DSPs of moderately using, instead of overusing, the online big data for maximized revenues.

Quality score (QS) plays a critical role in sponsored search advertising (SSA) auctions, and in practice is closely correlated to the historical click-through rate (CTR) of an advertisement. The CTR-QS correlation may impose great influence on advertisers' positioning strategies of selecting the targeting slots in the sponsored list. In the literature, however, QS is implicitly assumed to be an independent variable and exogenously assigned by Web search engines, so that little theoretical or managerial insights can be offered to help understand the positioning dynamics in SSA auctions with CTR-QS correlation. We strive to bridge this research gap in this paper. Based on a discrete time-dependent optimal control model, which explicitly captures the relationship between the historical CTR and QS, we determine the optimal strategy for revenue-maximizing advertisers' QS-based positioning decisions through a policy-iteration-based numerical approximation method. We also investigate two practically-used heuristic strategies, namely the greedy and farsighted positioning strategies, aiming to examine and help understand advertisers' real-world positioning dynamics. Our analysis indicates that both the optimal and greedy positioning strategies lead advertisers to monotonically increase or decrease their targeting slots over time, which may cause a polarization trend emerging in SSA markets. Meanwhile, the farsighted positioning strategy can accelerate the polarization. Our simulations show that both the greedy and farsighted strategies have good revenue performance. Our findings indicate that advertisers should monotonically adjust their targeting positions to maximize their revenue in CTR-QS correlated SSA auctions. Our findings also highlight the need for Web search engine companies to set a lowered weight for historical CTRs or use position-normalized CTRs in their QS measurements, so as to suppress the polarization trend.

This paper addresses issues related to the development of a computational theory and corresponding methods for studying complex socioeconomic systems. We propose a novel computational framework called ACP (Artificial societies, Computational experiments, and Parallel systems), targeting at creating an effective computational theory and developing a systematic methodological framework for socioeconomic studies. The basic idea behind the ACP approach is: 1) to model the complex socioeconomic systems as artificial societies using agent techniques in a "bottom-up” fashion; 2) to utilize innovative computing technologies and make computers as experimental laboratories for investigating socioeconomic problems; and 3) to achieve an effective management and control of the focal complex socioeconomic system through parallel executions between artificial and actual socioeconomic systems. An ACP-based experimental platform called MacroEconSim has been discussed, which can be used for modeling, analyzing, and experimenting on macroeconomic systems. A case study on economic inflation is also presented to illustrate the key research areas and algorithms integrated in this platform.