UnionPay (China)

Despite being a groundbreaking research that has achieved phenomenal success in various graph learning tasks, graph convolutional network (GCN) falters in link prediction (LP). This is primarily due to its inherent message-passing mechanism. On one hand, GCN has limited expressive power and cannot identify topological patterns such as counting 3-cycles, which is a cornerstone of LP heuristics. On the other hand, GCN restricts message flow to hardwired interactions, while in LP scenarios, many edges in the graph are missing. This contradiction hinders GCN from learning ideal node representations. To tackle these challenges, we propose TopoLink, a topology-enhanced graph Transformer for LP. We design a Simplified Graph Transformer that integrates classical GCN with a global attention module with linear complexity to simultaneously capture local and global information. Besides, we introduce extended persistent homology (EPH) to enrich the ability to detect topological features such as cycles. Specifically, to deeply understand abstract and complex topological information, we transform EPH into persistence images and, for the first time, apply computer vision techniques to refine them into vectorized topological fingerprints. Extensive experiments on real-world datasets demonstrate that TopoLink outperforms state-of-the-art methods with 0.55-18.84% improvements. The code is available at https://github.com/liulizhi1996/TopoLink.

Anti-money laundering (AML) is a classical data mining problem in finance applications. As well known, money laundering (ML) is critical to the effective operation of transnational and organized crime, which affects a country's economy, government, and social wellbeings. Financial services organizations facilitate the movement of money and have been enlisted by governments to assist with the detection and prevention of money laundering, which is a key tool in the fight to reduce crime and create sustainable economic development. In the application of AML, user identity and financial behavior data are widely used to detect laundering transactions. In recent years, an increasing number of money laundering activities have been conducted by organized criminal gangs while most existing works still treat the actions of each account as independent identity behavior without considering the group-level conspired interactions. Therefore, in this paper, we propose a group-aware deep graph learning-based approach for organized money-laundering detection. In particular, we design a community-centric encoder to represent the nodes and attributes in user transaction graphs and derive the adjacent gang behaviors. Then, we devise a scheme of local enhancement to accommodate nodes with similar transaction features, which are aggregated into gangs for downstream detection. Extensive experiments on the real-world dataset from one of the largest bank card alliances worldwide show that our proposed method outperforms state-of-the-art methods in both offline and online modes, showing the effectiveness of money laundering detection with group-aware deep graph learning.

PurposeIn the era of the digital economy, organizations are under much pressure to justify their information technology (IT) spending on digital transformation. Some organizations have thus implemented IT chargeback, an IT governance (ITG) mechanism, to clarify and allocate IT costs among various business units. While practitioners have stressed the importance of IT chargeback, there has been little theoretical effort that investigates its strategic effects and boundary conditions.Design/methodology/approachSynthesizing the ITG literature and the resource-based view (RBV), the authors develop a research model to examine if IT chargeback affects IT–business strategic alignment and, in turn, organizational performance and how human IT resources strengthen the impacts of IT chargeback. The authors designed a survey to collect data from 103 firms and tested the model using partial least squares (PLS).FindingsThe authors found that IT chargeback promoted strategic alignment and then organizational performance only for firms with business-competent chief information officers (CIOs), rather than IT-competent business executives.Originality/valueThis study enriches the ITG literature by exploring the strategic value of an IT cost governance mechanism (i.e. IT chargeback). This study further proposes and validates a measure of IT chargeback. Drawing on the RBV, this study quantitatively investigates the strategic impacts and boundary contingencies of IT chargeback. This study also advances the CIO literature by identifying the strategic leading role, instead of the traditional supporting role, of CIOs in modern organizations.

Spiking neural network (SNN) is a new generation of artificial neural network. It is more biologically plausible and has been widely concerned by scholars owing to its unique information coding schemes, rich spatiotemporal dynamics, and event-driven operating mode with low power. In recent years, SNN has been explored and applied in many fields such as medical health, industrial detection, and intelligent driving. First, the basic elements and learning algorithms of SNN are introduced, including classical spiking neuron models, spike-timing dependent plasticity (STDP), and common information coding methods. The advantages and disadvantages of the learning algorithms are also analyzed. Then, the mainstream software simulators and neuromorphic hardware of SNN are summarized. Subsequently, the research progress and application scenarios of SNN in terms of computer vision, natural language processing, and reasoning decision are introduced. Particularly, SNN has shown strong potentials in tasks such as object detection, action recognition, semantic cognition, and speech recognition, significantly improving computational performance. Future research and application of SNN should focus on strengthening the research on key core technologies, promoting the application of technological achievements, and continuously optimizing the industrial ecology, thus to catch up with the advanced international level. Moreover, continuous research and breakthroughs of brain-inspired systems and control theories will promote the establishment of large-scale SNN models and are expected to broaden the application prospect of artificial intelligence.

With the prevalence of mobile e-commerce, fraudulent transactions conducted by robots are becoming increasingly common in mobile payments, which is severely undermining market fairness and resulting in financial losses. It has become a difficult problem for mobile applications to identify robotic automation accurately and efficiently from a massive number of transactions. The current research does not propose any effective method or engineering implementation. In this article, an extension to boost algorithms is presented that permits the incorporation of prior human knowledge as a means of compensating for a training data shortage and improving prediction results. Prior human knowledge is accumulated from historical fraud transactions or transferred from different domains in the form of expert rules and blacklists. The knowledge is applied to extract risk features from transaction data, risk features together with normal features are input into the boosting algorithm to perform training, and therefore we incorporate boosting algorithm with prior human knowledge to improve the performance of the model. For the first time we verified the effectiveness of the method via a widely deployed mobile APP with 150+ million users, and by taking experiments on a certain dataset, the extended boosting model shows an accuracy increase from 0.9825 to 0.9871 and a recall rate increase from 0.888 to 0.948. We also investigated feature differences between robots and normal users and we discovered the behavior patterns of robotic automation that include less spatial motion detected by device sensors (1/10 of normal user pattern), higher IP group-clustering ratio (60% in robots vs. 15% in normal users), higher jailbroken device rate (92.47% vs. 4.64%), more irregular device names and fewer IP address changes. The quantitative analysis result is helpful for APP developers and service providers to understand and prevent fraudulent transactions from robotic automation.This article proposed an optimized boosting model, which has better use in the field of robotic automation detection of mobile phones. By combining prior knowledge and feature importance analysis, the model is more robust when the actual dataset is unbalanced or with few-short samples. The model is also more explainable as feature analysis is available which can be used for generating disposal rules in the actual fake mobile user blocking systems.

ABSTRACT Screen mirroring might be a way to improve older adults’ user experience of smart televisions (STVs) through smartphones. To examine this possibility, two experiments were conducted. Experiment I examined older adults’ difficulties of screen mirroring (mirroring smartphone screens to STVs) through five common touch gestures (“Drag,” “Slide,” “Zoom,” “Draw,” and “Handwrite”), in comparison to younger adults. The results indicated that a major problem for older adults is the frequent attention switching between the STV and smartphone screens. Therefore, experiment II explored how to reduce the need of attention switching through the touch gestures (“Tap,” “Slide + Tap,” and “Slide + Release”) and the button sizes (8, 14, and 20 mm) for different input postures. Thirty older adults participated in this experiment and their eye movements were tracked. Four major findings were derived. First, the “Zoom,” “Draw,” and “Handwrite” gestures in screen mirroring were difficult for older adults with a task completion rate lower than 68%. Second, the problem of frequent attention switching between the STV and smartphone was predominant for tapping tasks. Third, the “Slide + Tap” and “Slide + Release” touch gestures helped to reduce attention switching in tapping tasks more than the “Tap” for older adults, while the “Slide + Release” received the worst subjective feedback. Fourth, increasing the button size from 8 mm to 14 mm on smartphones can improve the task completion rate and the task efficiency in screen mirroring when older adults used the one-handed posture to tap.

Digital currency can reduce the issuance and circulation costs of physical cash and improve the convenience and transparency of economic activities. Therefore, digital currency has received widespread attention from central banks in recent years. As an important financial infrastructure, the UnionPay network is expected to provide effective support for the construction and operation of a digital currency system. This study uses the UnionPay network as basis to systematically propose a digital currency information system from three aspects, namely, digital currency account opening, exchange, and use, and further expounds on the operation mechanism of a digital currency information system prototype. This study also discusses how to apply blockchain technology to digital currency, such as using distributed ledger for digital currency confirmation and registration and using blockchain-enabled smart contracts to realize the forward guidance function of the central bank digital currency. Lastly, the current research introduces the value of using the UnionPay network to support the operation of a digital currency information system and its technical challenges, aiming to provide beneficial guidance and reference for future research effort.