As an increasing volume of international trade activities around the world, the amount of cross-boarder import declarations grows rapidly, resulting in an unprecedented scale of potentially fraudulent transactions, in particular false commodity code (e.g., HS Code). The incorrect HS Code will cause duty risk and adversely impact the revenue collection. Physical investigation by the customs administrations is impractical due to the substantial quantity of declarations. This paper provides an automatic approach by harnessing the power of machine learning techniques to relief the burden of customs targeting officers. We introduced a novel model based on the off-the-shelf embedding encoder to identify the correctness of HS Code without any human effort. Determining whether the HS Code is correctly matched with commodity description is a classification task, so the labelled data is typically required. However, the lack of gold standard labelled data sets in customs domain limits the development of supervised-based approach. Our model is developed by the unsupervised mechanism and trained on the unlabelled historical declaration records, which is robust and able to be smoothly adapted by the different customs administrations. Rather than typically classifying whether the HS Code is correct or not, our model predicts the score to indicate the degree of the HS Code being correct. We have evaluated our proposed model on the ground-truth data set provided by Dutch customs officers. Results show promising performance of 71% overall accuracy.

Comparing ontology matching systems are typically performed by comparing their average performances over multiple datasets. However, this paper examines the alignment systems using statistical inference since averaging is statistically unsafe and inappropriate. The statistical tests for comparison of two or multiple alignment systems are theoretically and empirically reviewed. For comparison of two systems, the Wilcoxon signed-rank and McNemar's mid-p and asymptotic tests are recommended due to their robustness and statistical safety in different circumstances. The Friedman and Quade tests with their corresponding post-hoc procedures are studied for comparison of multiple systems, and their [dis]advantages are discussed. The statistical methods are then applied to benchmark and multifarm tracks from the ontology matching evaluation initiative (OAEI) 2015 and their results are reported and visualized by critical difference diagrams.

Total variation a b s t r a c tThe l 1 -regularized least square problem has been considered in diverse fields.However, finding its solution is exacting as its objective function is not differentiable.In this paper, we propose a new one-layer neural network to find the optimal solution of the l 1 -regularized least squares problem.To solve the problem, we first convert it into a smooth quadratic minimization by splitting the desired variable into its positive and negative parts.Accordingly, a novel neural network is proposed to solve the resulting problem, which is guaranteed to converge to the solution of the problem.Furthermore, the rate of the convergence is dependent on a scaling parameter, not to the size of datasets.The proposed neural network is further adjusted to encompass the total variation regularization.Extensive experiments on the l 1 and total variation regularized problems illustrate the reasonable performance of the proposed neural network.

Ontology alignment is widely used to find the correspondences between different ontologies in diverse fields. After discovering the alignments, several performance scores are available to evaluate them. The scores typically require the identified alignment and a reference containing the underlying actual correspondences of the given ontologies. The current trend in the alignment evaluation is to put forward a new score (e.g., precision, weighted precision, semantic precision, etc.) and to compare various alignments by juxtaposing the obtained scores. However, it is substantially provocative to select one measure among others for comparison. On top of that, claiming if one system has a better performance than one another cannot be substantiated solely by comparing two scalars. In this article, we propose the statistical procedures that enable us to theoretically favor one system over one another. The McNemar’s test is the statistical means by which the comparison of two ontology alignment systems over one matching task is drawn. The test applies to a 2 × 2 contingency table, which can be constructed in two different ways based on the alignments, each of which has their own merits/pitfalls. The ways of the contingency table construction and various apposite statistics from the McNemar’s test are elaborated in minute detail. In the case of having more than two alignment systems for comparison, the family wise error rate is expected to happen. Thus, the ways of preventing such an error are also discussed. A directed graph visualizes the outcome of the McNemar’s test in the presence of multiple alignment systems. From this graph, it is readily understood if one system is better than one another or if their differences are imperceptible. The proposed statistical methodologies are applied to the systems participated in the OAEI 2016 anatomy track, and also compares several well-known similarity metrics for the same matching problem.

In global supply chains, a multitude of public and private organizations collaborate to reach the collective goal of transporting goods from the seller to the buyer. Given the dynamicity of global supply chains it is hard to predict which information is needed by whom at which point in time which often causes service delivery delays. Integrating relevant information, processes and services prevents disruptions in the supply chain and deterioration in service provisioning caused by missing information required for processes that need to be executed to supply services. In this paper, an ontology-based, event-driven architecture is designed for integrating information, processes and services that acts as a mechanism to coordinate service delivery in a global supply chain. The architecture is evaluated in the context of a global supply chain of plastic toys. This illustrates how the events facilitated by the architecture enable the availability of information in time which positively influences the delivery of a barge planning service.

To keep a network of enterprises sustainable, inter-organizational control measures are needed to detect or prevent opportunistic behaviour of network participants. We present a requirements engineering method for understanding control problems and designing solutions, based on an economic value perspective. The methodology employs a library of so-called control patterns, inspired by design patterns in software engineering. A control pattern is a generic solution for a common control problem. The usefulness and adequacy of the control patterns is demonstrated by a case study of the governance and control mechanisms of the Dutch public health insurance network for exceptional medical expenses (AWBZ).

There is an ongoing debate in law and accounting about the relative merits of principle-based versus rule-based regulatory systems. In this paper we characterize what kind of reasoning underlies the two styles of regulation. We adapt an original account of Verheij et al. (1998) to take aspects of the implementation context into account, such as the process of adoption of a new norm and the roles of the participants. The model is validated by a comparison between EU and US customs regulations intended to enhance safety and security in international trade. The EU regulations (AEO self-assessment) are essentially principle-based, whereas the American system (C-TPAT) is rule-based.

In the arena of international trade, multiple levels of governments (ranging from national to supranational) play an important role in regulating and controlling cross-border trade activities. Recently, they have also become powerful players in influencing decisions about inter-organizational systems (IOS). The influences of these multiple levels of governments on IOS are of an enormous scale and impact on businesses and national economies. Understanding them is a prerequisite for informed actions. From a theoretical point of view we contribute with the MLxMC framework, a conceptual framework, which combines a processual, multi-level approach with the motors of change. We use the MLxMC framework to explicitly highlight the influences of multiple levels of governments on IOS developments that take place in the highly regulated environment of international trade. We demonstrate how the framework can be applied to analyze such developments, extending the existing IOS research with models that explicitly acknowledge the role of government. The framework makes use of multi-level analysis by taking the political and institutional aspects into account. As an analytical tool, the framework can support business practitioners as well as policy-makers in their strategic choices of which level to engage at and with whom to collaborate in order to influence the debate.

Networked organizations, consisting of enterprises who exchange things of economic value with each other, often have participants who commit a fraud or perform other actions, not agreed in a contract. To explore such opportunistic behavior, and to design solutions to mitigate it, we propose the e3control approach. This approach takes the valuable objects, which are exchanged between enterprises, as a point of departure, and proposes a control patterns library to find solutions for various types of opportunistic behavior in network organizations. The practical use of the patterns is illustrated by a case study in the field of renewable electricity supply in UK.

Understanding consumer behaviour is of vital importance to consumer-oriented e-business models today. In this paper, we study the relationships between consumer perceptions of risk and trust and the attitude towards purchasing at a consumer-to-consumer electronic marketplace (EM). Typical for EM settings is that consumer behaviour is subject to perceptions of the selling party as well as of the institutional structures of the intermediary that is operating the EM. Building upon the well-established literature of trust, we consider the concepts of intermediary trust and seller trust. We extend this categorisation by introducing the concepts of intermediary risk and seller risk. We developed measurement instruments for intermediary risk and seller risk. All measurement scales have acceptable alphas and are unidimensional. An empirical study is conducted to explore the relationships between the risk and trust types and consumer purchase attitude. The results reveal significant, direct effects of seller trust and seller risk. Second-order effects of intermediary trust and intermediary risk are investigated and reported. The paper concludes with general observations and recommendations for research and practice.

Exploration and development of e-business models takes a series of viewpoints. One important perspective is the value web perspective, which can be modelled using the e3value methodology. However this perspective supposes a perfectly honest world, an assumption which turns out to be not true in practice. As a consequence, it is important to explore fraudulent behaviour of actors in value web, preferably by using a lightweight and graphical approach similar to the e3value methodology. Using a real-life e-business scenario (Bill of Lading) we show that e3value, with some extensions, is capable of modelling a control perspective that captures mechanisms for preventive and detective controls. Additionally, we show that such controls themselves can be seen as a kind of commercial services.

In this paper we introduce the DIagnostic and DEcision-theoretic framework for \nDEontic\nreasoning diO(de)2. The framework diO(de)2 formalizes reasoning about \nviolations and goals.\nWe use this framework to discuss two theories of normative reasoning, diagnosis \ntheory and\n(qualitative) decision theory. A crucial distinction between the two theories \nis their perspective\non time. Diagnosis theory reasons about incomplete knowledge and only considers \nthe past. It\ndistinguishes between violations and non-violations. Qualitative decision \ntheory reasons about\ndecision variables and considers the future. It distinguishes between fulfilled \nobligations and\nunfulfilled obligations. Moreover, we discuss the relation between the two \ntheories of normative\nreasoning and deontic logic. The theories formalize reasoning with norms, and \nthey are thus\ndifferent from deontic logic, that formalizes reasoning about norms.