In the business we are experiencing digital transformation by a higher speed of change and increasing complexity. Especially in the area of BPM this causesmoreprojectswhich fail. The reasons aremanifold butwell knownandpoint to the usage ofmore than 40 years old paradigms of software development. The gap between people formulation new requirements for processes and those creating the software for digitization and automation is getting larger. A solution is to involve business practitioners directly in programming. This disruptive approach is shifting the old software development paradigms and only possible if the basis for programming by businesspeople is based on subject-orientation and on a very simple and easy to use programming environment. Metasonic® Process Suite and Touch provides exactly this environment for coding the business logic of a process by businesspeople. The created process model serves both business and IT.Many examples realized on subject-oriented BPM prove this new concept pays off and is created big success. A comparison of TCO between S-BPM projects and projects using conventional approaches shows the financial advantages in more details. For BPM projects, using the S-BPM methodology and the metasonic® Process Suite & Touch yield significant time and cost savings. The savings are a direct result of the essential capabilities that set the S-BPM methodology and the metasonic® Process Suite & Touch apart from other approaches it focuses on subjects and their communication – the two key elements that are essential to any organization’s success.

Enterprise security is complicated by the use of mobile devices. These devices roam outside the protections of the enterprise core network. They operate closer to threats while simultaneously being farther from the enterprise, which makes compromise more likely and response more difficult. This paper describes an approach using software agents installed on endpoint devices to maintain security of these devices and their associated enterprise. These agents monitor local activity, prevent harmful behavior, allow remote management, and report back to the enterprise. The challenge in this environment is the security of the agents and their communication with the enterprise. This work presents an agent architecture that operates within a high-security Enterprise Level Security (ELS) architecture that preserves end-to-end integrity, encryption, and accountability. This architecture uses secure hardware for sensitive key operations and device attestation. Software agents leverage this hardware security to provide services consistent with the ELS framework. Additional agents leverage this baseline security to provide additional features and functions. This enables an enterprise to manage and secure all endpoint device agents and their communications with other enterprise services.

C-STEAMis a localized transdisciplinary education inChinese context with the goals of inheriting traditional culture and cultivating students’ STEAM literacy. It mainly has three core values: educational value of cultivating students’ core literacy; carrier value of inheriting outstanding traditional culture; and social value of booming regional culture. In order to see how a C-STEAM project works and effects, we adopted “6C” Instructional Design Model which is an operational steps including Contextual Experience, Connotation construction, Characteristic inquiry, Create artifact, Connect with society and Conclusion reflection. In a case study lasting 6 weeks on teaching Cantonese slang in a primary school in Foshan, China, questionnaires were used to analyze the students’ learning performance, STEAM literacy, cultural understanding and inheritance, and structured interviews were used to collect views and reflections of teachers and students. Results indicated that the application of “6C” Model has significantly improved students’ cultural understanding and cultural identity and cultural inheritance, and shown that C-STEAMplays an important role as the cultural carriers, brings some social value for constructing regional characteristic culture and enables to cultivate students’ core literacy and improves students’ transdisciplinary learning ability. More research is needed to ameliorate “6C” Model to achieve the core values of C-STEAM.

A pair of fully automatic brain tissue and tumor segmentation frameworks are introduced in current paper, these consist of a parallel and cascade architectures of a specialized convolutional deep neural network designed to develop binary segmentation. The main contributions of this proposal imply their ability to segment Magnetic Resonance Imaging (MRI) of the brain, of different acquisition modes without any parameter, they do not require any preprocessing stage to improve the quality of each slice. Experimental tests were developed considering BraTS 2017 database. The robustness and effectiveness of this proposal is verified by quantitative and qualitative results.

Moving target detection is widely used in the field of intelligent monitoring. This paper proposes an Improved Frame Difference Method Fusion Horn-Schunck Optical Flow Method (IFHS) algorithm to detect moving targets in video surveillance. First, a three-frame difference method is performed on the image to obtain a difference image. Second, the differential image is tracked according to the HS optical flow method to obtain an optical flow vector. Finally, the moving target is obtained by fusing the optical flow vector information of the two difference results. The problem of incomplete contours of moving targets detected in video surveillance by the separate optical flow method and the three-frame difference method is solved. The experimental simulation and experimental applications in different scenarios prove the robustness and accuracy of the method.

The DATA 2019 proceedings volume focuses on data management technologies and applications. The papers contribute to the understanding of relevant trends of current research and address such topics as: decision support systems, data analytics; data and Information quality, digital rights management.

In association rule mining, both the classical algorithms and today’s available tools either use binary data items or discretized data. However, in real-world scenarios, data are available in many different forms (numerical, text) and these types of data items are not supported in the classical association rule mining algorithms. There are some association rule mining algorithms that have been proposed for numerical data items but unfortunately, for working data scientists and decision makers, it is challenging to find concrete algorithms that fit their purposes best. Therefore, it is highly desired to have a study on the different existing numerical association rule mining algorithms (NARM). In this paper, we provide such a detailed study by thoroughly reviewing 24 NARM algorithms from different categories (optimization, discretization, distribution).

The problem of the learning of polynomial threshold units over a fixed set of polynomials is treated in the paper. We consider two approaches in the supervised learning: off-line relaxation-like algorithms and spectral off-line algorithms. Our research is focused on answering the question: what values of the learning rate provide the fast convergence of the learning algorithm? We propose and justify a new adaptive rule of the choice of learning rates, which ensures that on-line relaxation-like algorithms produce a desired weight vector of polynomial neural unit after a finite number of learning steps. Then, we deal with the spectral off-line learning algorithm and give the reasons that the extension of the range of acceptable values for the learning rate preserves the convergence and the finiteness of learning. The relaxation-like learning algorithm with margin is also considered. Its advantage is the finite learning time in the case when a constant learning rate belongs to (0, 2]. Next, we propose a new rule of the choice of initial approximation significantly accelerating the convergence both for on-line and for off-line learning. Finally, the simulation results are given for quadratic neural units, which confirm the validity and the advantage of the proposed approaches.

Unlike traditional single-PC applications, which have access to directly attached computational resources (CPUs, memory, and I/O devices), web applications have to deal with the resources scattered across the network. Besides, web applications are intended to be accessed by multiple users simultaneously. That not only requires a more sophisticated infrastructure but also brings new challenges to web application

The paper introduces the Secure kNN (SkNN) approach to data classification and querying. The approach is founded on the concept of Secure Chain Distance Matrices (SCDMs) whereby the classification and querying is entirely delegated to a third party data miner without sharing either the original dataset or individual queries. Privacy is maintained using two property preserving encryption schemes, a homomorphic encryption scheme and bespoke order preserving encryption scheme. The proposed solution provides advantages of: (i) preserving the data privacy of the parties involved, (ii) preserving the confidentiality of the data owner encryption key, (iii) hiding the query resolution process and (iv) providing for scalability with respect to alternative data mining algorithms and alternative collaborative data mining scenarios. The results indicate that the proposed solution is both efficient and effective whilst at the same time being secure against potential attack.

Even though, Massive MIMO (M-MIMO) technology offers good improvement in terms of performance and quality of communications between users and Base Stations (BS). This technology, still limited by a harmful constraint renowned Pilot Contamination (PC) problem. To beat this problem of PC, two approaches have been adapted as a medicine for this problem, which appears strongly in Multi-Cell M-MIMO systems. The first approach is based on assigning extra pilot sequences to different users, while the second approach obliges the reuse of the same set of pilots in different cells. This paper reviews, briefly, the problem of PC in M-MIMO systems. Thereafter, it analyzes and provides a comparison between two decontaminating strategies: the Soft Pilot Reuse (SPR) and the Pilot Assignment based on the Weighted Graph Coloring (WGC-PA), which are respectively based on the abovediscussed approaches. The analysis presented in this paper is focused on the uplink phase (i.e reverse link).

The importance of Treasury management, within a commercial bank has increased significantly over the last couple of years. After the 2008 financial crisis the role and responsibility of a Treasury department has changed in terms of scope and strategic importance, evolving from a transactional cash manager to the guardian of the balance sheet. In order tomeet this broader strategicmandate, Treasurers must therefore consider ways to become more effective and streamlined, while reducing time-consuming operational activities. Digitalisation can address many of the traditional Treasury challenges and provide a number of commercial and competitive benefits as well. However, to successfully adopt digital technologies and related digital innovations, Treasury requires a well-defined digital transformation plan. The Smart Digital TreasuryModel (SDTM)was developed to provide a comprehensive roadmap to assist a Treasury’s digital transition towards a next generation ‘smart’ Treasury department. This paper explores a key building block of the SDTM, which addresses the risks and threats that can arise from the adoption of new digital technology. The reason for focusing on this aspect is that many of the digital risks have no direct reference points with conventional banking activity or security measures. The result of this research is an approach that articulates Treasury specific digital risks and threats, as well as describes a risk management process that can be deployed as part of the digital transformation. The digital landscape is evolving the whole time; therefore, digital risk management activity in Treasury can’t be seen as a once-off exercise, but needs to evolve in line with market developments.

In the past decade, Artificial Intelligence (AI) has become a part of our daily lives due to major advances in Machine Learning (ML) techniques. In spite of an explosive growth in the raw AI technology and in consumer facing applications on the internet, its adoption in business applications has conspicuously lagged behind. For business/missioncritical systems, serious concerns about reliability and maintainability of AI applications remain. Due to the statistical nature of the output, software ‘defects’ are not well defined. Consequently, many traditional quality management techniques such as program debugging, static code analysis, functional testing, etc. have to be reevaluated. Beyond the correctness of an AI model, many other new quality attributes, such as fairness, robustness, explainability, transparency, etc. become important in delivering an AI system. The purpose of this paper is to present a view of a holistic quality management framework for ML applications based on the current advances and identify new areas of software engineering research to achieve a more trustworthy AI.

In the originally published version of the book there was an error in the first name of the second volume editor: "José Francisco Domínguez Mayo" should have been "Francisco José Domínguez Mayo". This has now been corrected.

Education is considered as a key factor for competitiveness on microas well as on macro-economic levels, i.e., for a single person, a company, or a country. Several industries have been identified with significant current and/or future workforce shortages. They include diverse areas, such as the health and the technology sectors. As a result, initiatives to foster STEM (science, technology, engineering and mathematics) have emerged in many countries. In this paper, we report on a project of Australian Catholic University and Munich University of Applied Sciences. These universities have developed a framework to support STEM education. The framework is based on R-Project, a leading free software environment that has gained increasing attention particularly in the field of data analysis (statistics, machine learning etc.) in the past decade. The framework is intended to address the following three main challenges in STEM education: mathematics and, in the field of technology, algorithmic programming and dynamic webpage development.

In a previous version of this publication, the affiliation of the second editor was incomplete. This has now been corrected.