Search

In this paper, we evaluate the Lbl2Vec approach for unsupervised text document classification. Lbl2Vec requires only a small number of keywords describing the respective classes to create semantic label representations. For classification, Lbl2Vec uses cosine similarities between label and document representations, but no annotation information. We show that Lbl2Vec significantly outperforms common unsupervised text classification approaches and a widely used zero-shot text classification approach. Furthermore, we show that using more precise keywords can significantly improve the classification results of similarity-based text classification approaches.

Aha moment is when users realize the value of using the software product, which is a key to driving revenue, mainly for B2B SaaS vendors. According to the Acquisition-Activation-Retention-Referral-Revenue (AARRR) model, the aha moment can refer to activation, and the following customer phase is retention. This research aims to find customers’ ”aha moment” through process mining. Since the customers in retention are obligated to experience activation before, the research first identifies milestone actions in terms of product features and user roles to cluster the retention customers. Evaluation is performed through the data of the clustered customers from the time before they move to the retention phase. The event log analysis is discussed based on multiple dimensions: product solution, time, and user roles. The research mainly applies the process mining technique, heuristic miner, to discover the customer’s behavior patterns. Apart from marketing funnels, Moreover concept of human-computer interaction is focused on event classification and data cleaning, which is practical for cleaning UI logs. The discovered processes and aha moment can guide future product development and value proposition re-stratigizing.

Enterprises always do their business within some restrictions. In a team of enterprise architects, the restrictions are transformed into the modelling conventions and the corresponding modelling constraints that should be consistently applied across all enterprise models. This paper presents an approach for refining and formali\-zing modeling conventions into modelling constraints and using them for assessment of enterprise models by a software component called ArchiChecker. The specifics of the proposed approach is that the modeling conventions are first visualized and formalized using the types of elements and relationships of the ArchiMate modeling language, that is also used for modelling of enterprise views. The ArchiMate elements and relationships serve as types to formulate constraints. The elements and relationships in an ArchiMate model are instances of the ArchiMate elements and relationships. Using these types and instances the ArchiChecker automatically generates the lists of violations of modeling conventions in the enterprise models. Each violation shows how a specific enterprise view deviates from a given modeling convention. The paper reports a case study of application of the proposed approach to enterprise modelling views and modelling conventions used in a medical center. The case study is used to discuss the added value of formalization and automated assessment of modelling constraints in enterprise modelling.

CRM platforms heavily depend on high-quality data, where poor-quality data can negatively influence its adoption. Additionally, these platforms are increasingly interconnected and complex to meet growing needs of customers. Hence, delivery and consultancy of CRM platforms becomes highly complex. In this study, we propose a CRM data quality management framework that supports CRM delivery and consultancy firms to improve data quality management practices within their projects. The framework should also improve data quality within CRM solutions for their clients. We extract best practices for CRM data quality management by means of a literature study on data quality definition and measurement, data quality challenges, and data quality management methods. In a case study at an IT consultancy company, we investigate how CRM delivery and consultancy projects can benefit from the incorporation of data quality management practices. The design of the framework is validated by means of confirmatory focus groups and a questionnaire. The results translate into a framework that provides a high-level overview of data quality management practices incorporated in CRM delivery and consultancy projects. It includes the following components: Client profiling, project definition, preparation, migration/integration, data quality definition, assessment, and improvement.

Process mining derives knowledge of the execution of processes through analyzing behavior as observed from real-life events. While benefits of process mining are widely acknowledged, finding an adequate level of detail at which a mined process model is suitable for a specific stakeholder is still an ongoing challenge. Process models can be mined at different levels of abstraction, often resulting in either highly complex or highly abstract process models. This may have an important impact on the comprehensibility of the process model, which can also differ from the perspective of a particular stakeholder. To address this problem from a stakeholder-centric perspective, we propose a methodology for determining an appropriate level of process model abstraction. To this end, we use quantitative metrics on process models as well as a qualitative evaluation by using a technology acceptance model (TAM). A logistics case study involving the fuzzy process mining discovery algorithm shows init ial evidence that the use of appropriate abstraction levels is key when considering the needs of various stakeholders.

CRM platforms heavily depend on high-quality data, where poor-quality data can negatively influence its adoption. Additionally, these platforms are increasingly interconnected and complex to meet growing needs of customers. Hence, delivery and consultancy of CRM platforms becomes highly complex. In this study, we propose a CRM data quality management framework that supports CRM delivery and consultancy firms to improve data quality management practices within their projects. The framework should also improve data quality within CRM solutions for their clients. We extract best practices for CRM data quality management by means of a literature study on data quality definition and measurement, data quality challenges, and data quality management methods. In a case study at an IT consultancy company, we investigate how CRM delivery and consultancy projects can benefit from the incorporation of data quality management practices. The design of the framework is validated by means of confirmatory focus groups and a questionnaire. The results translate into a framework that provides a high-level overview of data quality management practices incorporated in CRM delivery and consultancy projects. It includes the following components: Client profiling, project definition, preparation, migration/integration, data quality definition, assessment, and improvement.

To measure Data Governance, researchers designed a few Data Governance Maturity Models, but the used capabilities are very different, resulting in different measurement outcomes. This research aims to find a substantiated and validated set of Data Governance capabilities for Data Governance Maturity Models design. We apply the Maturity Model design procedure model of Becker et al. for methodology, which we complement with the Generic Capability Reference model to validate the capabilities. As results, we find a proper set of Data Governance capabilities for designing a Data Governance Maturity Models. Furthermore, we validate the set of DG capabilities against the GCR model, of which we conclude that the Generic Capability Reference model is valid as a reference model for the (re)design of Maturity Models.

Scientific datasets are increasingly stored, published, and re-used online. This has prompted major search engines to start services dedicated to finding research datasets online. However, to date such services are limited to keyword search, and provide little or no semantic guidance. Determining the scientific domain for a given dataset is a crucial part in dataset recommendation and search: "Which research domain does this dataset belong to?". In this paper we investigate and compare a number of novel ontology-based methods to answer that question, using the distance between a domain-ontology and a dataset as an estimator for the domain(s) into which the dataset should be classified. We also define a simple keyword-based classifier based on the Normalized Google Distance, and we evaluate all classifiers on a hand-constructed gold standard. Our two main findings are that the seemingly simple task of determining the domain(s) of a dataset is surprisingly much harder than expected (even when performed under highly simplified circumstances), and that (again surprisingly), the use of ontologies seems to be of little help in this task, with the simple keyword-based classifier outperforming every ontology-based classifier. We constructed a gold-standard benchmark for our experiments which we make available online for others to use.

As knowledge graphs are getting increasingly adopted, the question of how to maintain the validity and accuracy of our knowledge becomes ever more relevant. We introduce context-aware constraints as a means to help preserve knowledge integrity. Context-aware constraints offer a more fine-grained control of the domain onto which we impose restrictions. We also introduce a bottom-up anytime algorithm to discover contextaware constraint directly from heterogeneous knowledge graphs-graphs made up from entities and literals of various (data) types which are linked using various relations. Our method is embarrassingly parallel and can exploit prior knowledge in the form of schemas to reduce computation time. We demonstrate our method on three different datasets and evaluate its effectiveness by letting experts on knowledge validation and management assess candidate constraints in a real-world knowledge validation use case. Our results show that overall, context-aware constraints are to an extent useful for knowledge validation tasks, and that the majority of the generated constraints are well balanced with respect to complexity.

Entity resolution in databases focuses on detecting and merging entities that refer to the same real-world object. Collective resolution is among the most prominent mechanisms suggested to address this challenge since the resolution decisions are not made independently, but are based on the available relationships within the data. In this paper, we introduce a novel resolution approach that combines the essence of collective resolution with rules and transformations among entity attributes and values. We illustrate how the approach’s parameters are optimized based on a global optimization algorithm, i.e., simulated annealing, and explain how this optimization is performed using a small training set. The quality of the approach is verified through an extensive experimental evaluation with 40M real-world scientific entities from the Patstat database.

Organizations across the world today face similar problems, thus, solutions to these problems could also be re-used across many organizations. Climate change is one such common problem to organizations, where any solution implemented by one could be re-used by many others. These re-usable solutions could be Enterprise Architecture Patterns, but there is a lack of guidance on how to use them in practice. This study proposes an extension to the commonly used TOGAF, to better leverage these reusable solutions. Thus, this enables organizations using this framework to enhance it with Enterprise Architecture Patterns. As part of the proposed methodology, re-usable patterns supporting sustainable characteristics are built. The resulting methodology is validated with an expert panel, gathering positive comments.

Contains fulltext : 233865.pdf (Publisher’s version ) (Closed access) ICEIS 2021

Chaos emerges with the ever growing amounts of data and information within organisations. But it is problematic to manage these valuable assets and also remain accountable and compliant for them because there is no agreement about even their definitions. Our objective is to propose a coherent set of definitions for data governance and information governance within and across organisations in relation with data and information as underlying concepts. As a research method, we explore elements from existing definitions in literature about the Data-Information-Knowledge-Wisdom pyramid and about data governance and information governance. Classification of these elements and coding them in concepts during discussions among peers resulted in a new vocabulary. This forms the basis for formulation and design of an original coherent set of definitions for data, information, meaning, data governance and information governance. This research is grounded, goal oriented and uses multiple accepted literature review methods. But it is limited to the literature found and the IS domain.

Organizations around the world today face many challenges, and not all of them are unique. Such repeatable problems can themselves be solved by reusable solutions. This study focuses on Enterprise Architecture Patterns, reusable solutions to repeating challenges organizations face. Through a systematic literature review, we aim to gather an exhaustive recollection of all such patterns in literature. In order to be as comprehensive as possible, the sample included related fields also researching patterns. The result is a collection of 593 patterns and their respective analysis, which will help practitioners tackle their challenges.

We propose in this paper a conceptual framework for the hierarchical decomposition of Non-Functional Requirements (NFRs) from the business process level to the web service level. This framework seeks to reduce the dependence on a particular IT expert’s knowledge by simplifying the dialog between the business and IT areas. The proposed framework relies on a structure of NFRs interdependence. The main reference was the ISO/IEC 25010 Product Quality Model, extended by additional software quality models and particular QoS attributes. This framework is accompanied by an extensive dictionary of non-functional requirements for both business processes and web services that can serve as a reference for researchers and industry practitioners. We assume that orchestrating web services to run business processes requires a rigorous definition of the functional requirements and NFRs of these web services. Web service NFRs are often defined as Quality of Service (QoS) attributes, which is done at the implementation level by IT teams. The definition of QoS attributes should consider the business process NFRs, since misinterpretations of web service NFRs may affect the behavior of the web services and hence achieving the business goals. The approaches proposed so far in the literature are still heavily dependent on an IT expert’s knowledge to identify the appropriate QoS attributes required to meet particular business process NFRs. However, defining appropriate QoS attributes without reference to business process-level NFRs may be a costly, time-consuming task.

We propose in this paper a conceptual framework for the hierarchical decomposition of Non-Functional Requirements (NFRs) from the business process level to the web service level. This framework seeks to reduce the dependence on a particular IT expert’s knowledge by simplifying the dialog between the business and IT areas. The proposed framework relies on a structure of NFRs interdependence. The main reference was the ISO/IEC 25010 Product Quality Model, extended by additional software quality models and particular QoS attributes. This framework is accompanied by an extensive dictionary of non-functional requirements for both business processes and web services that can serve as a reference for researchers and industry practitioners. We assume that orchestrating web services to run business processes requires a rigorous definition of the functional requirements and NFRs of these web services. Web service NFRs are often defined as Quality of Service (QoS) attributes, which is done at the implementation level by IT teams. The definition of QoS attributes should consider the business process NFRs, since misinterpretations of web service NFRs may affect the behavior of the web services and hence achieving the business goals. The approaches proposed so far in the literature are still heavily dependent on an IT expert’s knowledge to identify the appropriate QoS attributes required to meet particular business process NFRs. However, defining appropriate QoS attributes without reference to business process-level NFRs may be a costly, time-consuming task.

The analysis of a historical document of a registry office from nineteenth century through Raman spectroscopy was performed by comparing with an ensemble of standard samples synthesized from cotton and linen raps. Structural parameters of the fibers of the cellulose, such as crystallinity, chain length, intermolecular interactions and packing, were obtained by the measurement of intensity ratios of some marker features. The results suggested the historical document has a cellulosic support originated from cotton, probably produced without basic pH treatment of the raw material.

Business monitoring approaches usually address indicators associated with processes only at the service level; i.e., related to the services implementing the processes. Monitoring at the service level raises technical measures geared to Information Technology (IT) managers. Monitoring of Key Performance Indicators (KPIs) is usually carried out at a higher level, but transversely to the organization’s processes, i.e., uncoupled from the processes. We present a component designed to aid in strategic alignment between business and IT by monitoring Non-Functional Requirements (NFR) of processes based on Quality of Service attributes. This component aims to allow business managers to monitor process executions by focusing on the indicators that truly respond to the execution of such processes. We evaluated the component via a proof of concept.

Business monitoring approaches usually address indicators associated with processes only at the service level; i.e., related to the services implementing the processes. Monitoring at the service level raises technical measures geared to Information Technology (IT) managers. Monitoring of Key Performance Indicators (KPIs) is usually carried out at a higher level, but transversely to the organization’s processes, i.e., uncoupled from the processes. We present a component designed to aid in strategic alignment between business and IT by monitoring Non-Functional Requirements (NFR) of processes based on Quality of Service attributes. This component aims to allow business managers to monitor process executions by focusing on the indicators that truly respond to the execution of such processes. We evaluated the component via a proof of concept.

Non-functional Requirements (NFRs) of web services are defined by IT teams at the implementation level often as Quality of Service (QoS) attributes. Orchestrating web services to run business processes requires a rigorous definition of the NFRs of such web services. The definition of QoS attributes should consider the business process NFRs since misinterpretations of web service NFRs may affect the behavior of the web services and hence achieving the business goals. The approaches proposed so far are still heavily dependent on an IT expert’s knowledge to identify the appropriate QoS attributes required to meet particular business process NFRs. Defining appropriate QoS attributes without reference to business process-level NFRs may be a costly, time-consuming task. We propose a conceptual framework for the hierarchical decomposition of NFRs from the business process level to the web service level. This framework seeks to reduce the dependence on a particular IT expert’s knowledge by simplifying the dialog between the business and IT areas. The proposed framework relies on a structure of NFRs interdependence. The main reference was the ISO/IEC 25010 Product Quality Model, extended by additional software quality models and particular QoS attributes.

Non-functional Requirements (NFRs) of web services are defined by IT teams at the implementation level often as Quality of Service (QoS) attributes. Orchestrating web services to run business processes requires a rigorous definition of the NFRs of such web services. The definition of QoS attributes should consider the business process NFRs since misinterpretations of web service NFRs may affect the behavior of the web services and hence achieving the business goals. The approaches proposed so far are still heavily dependent on an IT expert’s knowledge to identify the appropriate QoS attributes required to meet particular business process NFRs. Defining appropriate QoS attributes without reference to business process-level NFRs may be a costly, time-consuming task. We propose a conceptual framework for the hierarchical decomposition of NFRs from the business process level to the web service level. This framework seeks to reduce the dependence on a particular IT expert’s knowledge by simplifying the dialog between the business and IT areas. The proposed framework relies on a structure of NFRs interdependence. The main reference was the ISO/IEC 25010 Product Quality Model, extended by additional software quality models and particular QoS attributes.

Genipa americana, pertencente à família Rubiaceae, árvore de grande porte que é encontrada em várias áreas do Brasil, popularmente conhecida como Jenipapeiro sendo utilizada para no combate à doenças hepáticas e inflamatórias, além de reduzir o colesterol. Este estudo tem como principal objetivo avaliar a prospecção fitoquímica, antibacteriana e modulatória do extrato hidroalcoólico da casca do caule de Genipa americana frente a cepas de bactérias padrões e multirresistentes. Para a análise da atividade antibacteriana do extrato hidroalcoólico, foi realizado o teste de microdiluição em caldo para determinação da concentração inibitória mínima (CIM), e a modulação de antimicrobianos por meio de gentamicina, amicacina e clindamicina. Os resultados obtidos da CIM pelas bactérias Escherichia coli, Staphylococcus aureus e Pseudomonas aeruginosa foram ≥ 1024µg/mL. Na modulação modificação da concentração de dois antibióticos para cada cepa bacteriana. Assim, é fato rematar que este extrato tem notável atividade moduladora principalmente quando testados contra bactérias de interesse clínico como a P. aeruginosa, bactérias oportunista, ressaltando que deve-se ser realizados mais estudo sobre o potencial desta espécie, que possui compostos fitoquímicos importantes capazes de reduzir a resistência a antimicrobianos.

As data analysis techniques progress, the focus shifts from simple tabular data to more complex data at the level of business objects. Therefore, the evaluation of such data analysis techniques is far from trivial. However, due to confidentiality, most researchers are facing problems collecting available real data to evaluate their techniques. One alternative approach is to use synthetic data instead of real data, which leads to unconvincing results. In this paper, we propose a framework to automatically operate information systems (supporting operational processes) to generate semi-real data (i.e., “operations related data” exclusive of images, sound, video, etc.). This data have the same structure as the real data and are more realistic than traditional simulated data. A plugin is implemented to realize the framework for automatic data generation.

The physical environment affects individuals emotionally, behaviorally and cognitively. Servicescapes or atmospherics studies the effect of environmental stimulation to consumers. Environmental stimuli affect consumers’ attention and cause them emotions such as pleasure and arousal; those emotional responses affect in turn consumers’ behavioral responses such us approach and avoidance tendencies towards the environment. Hence, the level of arousal-nonarousal and the pleasure-displeasure experienced by a consumer along with other intervening variables such as momentary mood and stimulus screening ability will determine his/her approach-avoidance responses towards the environmental stimuli. The paper studies atmospherics with neuromarketing and conventional marketing research methods: Specifically, using electroencephalography (EEG) and surveys, it focuses on the effect of background music on Consumer’s arousal, pleasure, attention and approach/avoidance tendencies. The results are expected to have significant academic relevance both for the servicescapes / atmospherics and the neuromarketing / consumer neuroscience research streams of literature and also managerial implications.

A 3D entangled coordination polymer {[Co4(glu)4(bpp)2]}n 1 was synthesized from a solvothermal reaction between Co(II), glutaric acid (H2glu) and 1,3-bis(4-pyridyl)propane (bpp). The crystal structure was determined by single crystal X-ray diffraction and the topological description indicates a 3D → 3D interpenetration in the pcu-coordination polymer, pillared by bpp ligands. Crystal structure presents four crystallographically distinct Co(II) metal centers conferring different coordination environments: tetrahedral, octahedral and trigonal bipyramidal. Spectroscopic (FTIR and Raman), thermal analysis and magnetic studies were employed to chemical characterization. SQUID magnetometry analysis indicated weak overall antiferromagnetic interactions between the Co(II) ions. High-Field Electron Paramagnetic Resonance (HFEPR) spectra suggest that the Co(II) ions in compound 1 are high-spin (S = 3/2). The material showed good thermal stability decomposing above 300 °C. CO2 adsorption studies carried out in the pressure range of 0 to 40 bar at room temperature revealed a favorable solid-gas interaction at low pressure. The amount of CO2 adsorbed on compound 1 continuously increased as the pressure increased and no gas saturation is observed suggesting a multilayer phenomenon of CO2 uptake. Powder X-ray diffraction analysis of compound 1 after CO2 measurements at 40 bar revealed no structural change.

This paper aims to map issues (risks and problems) at the interface of agile and traditional development approaches in hybrid organizations which have an impact on coordination and cooperation. Successfully combining agile and traditional development methods appears to be quite a challenge for many hybrid organizations. Both methods have their own strengths and added value but also bring their own culture and conditions. Combining these can lead to problems. If we want to handle such problems, we first need to understand the issues that can cause such problems. This study is aimed at identifying and validating an overview of these issues. Based on an exploration of literature a preliminary overview of issues was derived. These were classified into a coherent set. The result was validated in a case study within a large financial institute in the Netherlands. The resulting list of twenty-four issues can be used as a starting point for handling the problem area.

Software developing organizations nowadays have a wide choice when it comes to sourcing software components. This choice ranges from developing or adapting in-house developed components via buying closed source components to utilizing open source components. This study aims at structured decision support in this type of decision. As a basis for this study an initial set of criteria is taken, that has been identified and validated in a particular software development environment in a previous study on the subject (Kusters et al, 2016). In the paper at hand we report on the results of the application and validation of the initial set of sourcing criteria in a completely different case study environment, namely a software environment in that. medical embedded software is being developed. In addition, and based on the outcomes of our case study, a further step is made towards structured decision support for sourcing decisions by the development of an initial sourcing criteria framework.

The service time of container vessels is the main indicator of the competitiveness of a maritime container terminal. Vessels have to be berthed along the quay, a subset of quay cranes must be assigned to them and work schedules have to be planned for unloading the import containers and loading the export containers. This work addresses the Tactical Berth Allocation Problem, in which the vessels are assigned to a given berth, and the Quay Crane Scheduling Problem, for which the work schedules of the quay cranes are determined. The nature of this environment gives rise to inaccurate knowledge about the information related to the incoming vessels. Therefore, the aforementioned optimization problems can be tackled by considering fuzzy arrival times for the vessels and fuzzy processing times for the loading/unloading operations. Two fuzzy mathematical models are provided to solve the problems at hand. The computational experiments carried out in this work corroborate the effectiveness of the proposed methodologies.

A eletroforese de proteínas plasmáticas é um fator chave na prática clínica médico-veterinária dos Centros de Recuperação de Animais Selvagens. Naqueles animais onde a história clínica raramente está disponível, o uso desta técnica é um excelente auxílio para a diferenciação entre processos inflamatórios agudos e crónicos. Embora existam vários métodos de eletroforese de proteínas plasmáticas, a eletroforese capilar de zona oferece uma melhor resolução para a interpretação do proteinograma do que as técnicas baseadas em gel. A Coruja-do-mato (Strix aluco) é uma ave de rapina noturna comum em toda a Europa. Esta é uma das espécies mais comummente admitida em centros de recuperação de animais selvagem. O objetivo principal deste trabalho é a definição de valores de referência para as diferentes frações das proteínas plasmáticas, por eletroforese capilar, nesta espécie. Amostras de sangue heparinizadas foram colhidas de 46 Corujas-do-mato saudáveis admitidas no Centro de Recuperação de Animais Selvagens do Hospital Veterinário da Universidade de Trás-os-Montes e Alto Douro, entre janeiro de 2015 e julho de 2017, e analisadas por eletroforese capilar de zona, de forma a obter intervalos de referência para as proteínas plasmáticas, desta espécie em Portugal. Este estudo seguiu as diretrizes da Sociedade Americana de Patologia Clínica Veterinária para a determinação de intervalos de referência, em animais. Métodos de estatística robusta foram utilizados para calcular intervalos de referência para as proteínas totais (2,8-4,3 g/dl), albumina (1,2-2,4 g/dl), α2-globulinas (0,1-1,2 g/dl), β-globulinas (0,1-1,4), γ-globulinas (0-0,6 g/dl) e rácio A/G (0,6-1,9). O intervalo de referência para a fração das α1-globulinas (0-0,5 g/dl) foi calculado por métodos não paramétricos. Uma diferença estaticamente significativa foi encontrada entre os resultados de machos e fêmeas na fração das α1-globulinas. A idade também influenciou o intervalo de referência para a fração γ-globulinas, sendo significativamente diferente entre ninhadas e filhotes ou adultos. Os intervalos de referência obtidos neste estudo foram comparados com os existentes na literatura, para a mesma espécie e podem ser utilizados como guia na avaliação do estado de saúde de Corujas-do-mato selvagens, admitidas em centros de recuperação de animais selvagens, em Portugal.

The Scaled Agile Framework (SAFe) has emerged over the last years as an approach which supports the improvement of software and systems development. SAFe claimes solutions for business challenges, such as shortening cycle's times, improving product quality, increasing team members' satisfaction, and involving the customer in product development. However, regarding customer involvement, there is limited research, both in SAFe and in real-life agile software development projects. In previous work we developed an initial conceptual customer involvement model for the SAFe domain in Philips Medical Systems. In this paper this initial model will be extended and enriched on the basis of a case study in an industrial company.

In this paper, we discuss how to apply an autoencoder to detect anomalies in payment data derived from an Real-Time Gross Settlement system. Moreover, we introduce a drill-down procedure to measure the extent to which the inflow or outflow of a particular bank explains an anomaly. Experimental results on real-world payment data show that our method can detect the liquidity problems of a bank when it was subject to a bank run with reasonable accuracy.

We discuss how an autoencoder can detect system-level anomalies in a real-time gross settlement system by reconstructing a set of liquidity vectors. A liquidity vector is an aggregated representation of the underlying payment network of a settlement system for a particular time interval. Furthermore, we evaluate the performance of two autoencoders on real-world payment data extracted from the TARGET2 settlement system. We do this by generating different types of artificial bank runs in the data and determining how the autoencoders respond. Our experimental results show that the autoencoders are able to detect unexpected changes in the liquidity flows between banks.

It is known that the comprehension of spatial prepositions involves the deployment of visual attention. For example, consider the sentence “The salt is to the left of the stove”. Researchers [29, 30] have theorized that people must shift their attention from the stove (the reference object, RO) to the salt (the located object, LO) in order to com- prehend the sentence. Such a shift was also implicitly assumed in the Attentional Vector Sum (AVS) model by [35], a cognitive model that computes an acceptability rating for a spatial preposition given a display that contains an RO and an LO. However, recent empirical findings showed that a shift from the RO to the LO is not necessary to understand a spatial preposition ( [3], see also [15, 38]). In contrast, these findings suggest that people perform a shift in the reverse direction (i.e., from the LO to the RO). Thus, we propose the reversed AVS (rAVS) model, a modified version of the AVS model in which attention shifts from the LO to the RO. We assessed the AVS and the rAVS model on the data from [35] using three model simulation methods. Our simulations show that the rAVS model performs as well as the AVS model on these data while it also integrates the recent empirical findings. Moreover, the rAVS model achieves its good performance while being less flexible than the AVS model. (This article is an updated and extended version of the paper [23] presented at the 8th International Conference on Agents and Artificial Intelligence in Rome, Italy. The authors would like to thank Holger Schultheis for helpful discussions about the additional model simulation.)

It is acknowledged that knowledge management (KM) brings several benefits for an organization, and two types of knowledge management strategies exist: Codification and personalization, and a gap exisits between both strategies. In this paper, the context of KM is placed on the campus of higher education. A knowledge capturing tool, in the form of a tablet based APP, is introduced to capture problem solving knowledge during a student project meeting. This tool intends to propose a new way of technology support to codify knowledge during socialization. A series of comparative experiments were undertaken to investigate the APP's influence on users' meeting experiences. This paper will present results based on interviews with the participants, which showed both positive and negative effects of the APP usage on their meeting experience.

This paper focuses on cyber-security simulations in networks modeled as a Markov game with incomplete information and stochastic elements. The resulting game is an adversarial sequential decision making problem played with two agents, the attacker and defender. The two agents pit one reinforcement learning technique, like neural networks, Monte Carlo learning and Q-learning, against each other and examine their effectiveness against learning opponents. The results showed that Monte Carlo learning with the Softmax exploration strategy is most effective in performing the defender role and also for learning attacking strategies.

Most of the Brazilian universities employ teaching-learning methodologies based on classic frontal lectures. The Medicine Programme of the Federal University of S˜ao Carlos (UFSCar) is an exception, since it employs active learning methodologies. The Educational and Academic Management System for Courses Based on Active Learning Methodologies (EAMS-CBALM) was built and it is currently used to support this programme, and has been made available for other programmes as well. This system was developed using Scrum, but during its development project it was often necessary to reconsider system behaviour scenarios, and consequently the product backlog items, mainly due to poor communication between the Product Owner (PO) and the development team. This paper discusses a case study in which Behaviour-Driven Development (BDD) has been used in combination with Scrum to redesign some EAMS-CBALM components. The paper demonstrates that the communication between the PO and the development team can be improved by using BDD as a communication platform to unambiguously define system requirements and automatically generate test suites.

In this paper, we discuss how to apply an autoencoder to detect anomalies in payment data derived from an Real-Time Gross Settlement system. Moreover, we introduce a drill-down procedure to measure the extent to which the inflow or outflow of a particular bank explains an anomaly. Experimental results on real-world payment data show that our method can detect the liquidity problems of a bank when it was subject to a bank run with reasonable accuracy.

International audience; Understanding the dynamical behaviour of cellular systems requires the development of effective modelling techniques. The modeling aims to facilitate the study and understanding of the dynamic behaviour of these systems, by the simulation of their designed models. Complex biomolecular networks are the basis of these models. In this paper, we propose a method of qualitative reasoning, based on a formal logical modeling, to qualitatively simulate the biomolecular network and interpret it behaviour over time. The power of our approach is illustrated by applying it to the case study of the autoregulation of the bacteriophage T4 gene 32.

The importance of web-based knowledge communities (WKCs) in the 'network society' is growing. This trend is seen in many disciplines, like education, government, finance and other profit- and non-profit organisations. There is a need for understanding the development of these online communities in order to steer it and to affect the impact it has. In this research, we aimed to identify interaction patterns in these communities to visualize and understand community developments, and show the relevance of WKCs for the development of learning education. We conducted a content analysis and a network analysis on big social data to identify the patterns in two Facebook-groups which were focused on educational development. Analysis of interaction patterns enabled us to identify three interaction stages within WKCs in educational settings: introduction, evolution and maturity. In the first stage, participants mainly introduce themselves. In the second stage, one shares information and in the final stage, participants are more open to share their opinions. The study shows that our network analysis approach is appropriate to analyse and visualize the development of interaction patterns and the results could help us to steer communities effectively and efficiently.

We present a bootstrapping algorithm to create a semantic lexicon from a list of seed words and a corpus that was mined from the web. We exploit extraction patterns to bootstrap the lexicon and use collocation statistics to dynamically score new lexicon entries. Extraction patterns are subsequently scored by calculating the conditional probability in relation to a non-related text corpus. We find that verbs that are highly domain related achieved the highest accuracy and collocation statistics affect the accuracy positively and negatively during the bootstrapping runs.

Regier and Carlson (2001) have investigated the processing of spatial prepositions and developed a cognitive model that formalizes how spatial prepositions are evaluated against depicted spatial relations between objects. In their Attentional Vector Sum (AVS) model, a population of vectors is weighted with visual attention, rooted at the reference object and pointing to the located object. The deviation of the vector sum from a reference direction is then used to evaluate the goodness-of-fit of the spatial preposition. Crucially, the AVS model assumes a shift of attention from the reference object to the located object. The direction of this shift has been challenged by recent psycholinguistic and neuroscientific findings. We propose a modified version of the AVS model (the rAVS model) that integrates these findings. In the rAVS model, attention shifts from the located object to the reference object in contrast to the attentional shift from the reference object to the located object implemented in the AVS model. Our model simulations show that the rAVS model accounts for both the data that inspired the AVS model and the most recent findings.

Self-Organizing Map (SOM) is a neural network model which is used to obtain a topology-preserving mapping from the (usually high dimensional) input/feature space to an output/map space of fewer dimensions (usually two or three in order to facilitate visualization). Neurons in the output space are connected with each other but this structure remains fixed throughout training and learning is achieved through the updating of neuron reference vectors in feature space. Despite the fact that growing variants of SOM overcome the fixed structure limitation they increase computational cost and also do not allow the removal of a neuron after its introduction. In this paper, a variant of SOM is proposed called AMSOM (Adaptive Moving Self-Organizing Map) that on the one hand creates a more flexible structure where neuron positions are dynamically altered during training and on the other hand tackles the drawback of having a predefined grid by allowing neuron addition and/or removal during training. Experiments using multiple literature datasets show that the proposed method improves training performance of SOM, leads to a better visualization of the input dataset and provides a framework for determining the optimal number and structure of neurons., Comment: ICAART 2016 accepted full paper

A three-day asynchronous, interactive workshop was held at ICE-B’10 in Piraeus, Greece in July of 2010. This event captured conference themes for e-Business challenges and directions across four subject areas: a) e-Business applications and models, b) enterprise engineering, c) mobility, d) business collaboration and e-Services, and e) technology platforms. Quality Function Deployment (QFD) methods were used to gather, organize and evaluate themes and their ratings. This paper summarizes the most important themes rated by participants: a) Since technology is becoming more economic and social in nature, more agile and context-based application develop methods are needed. b) Enterprise engineering approaches are needed to support the design of systems that can evolve with changing stakeholder needs. c) The digital native groundswell requires changes to business models, operations, and systems to support Prosumers. d) Intelligence and interoperability are needed to address Prosumer activity and their highly customized product purchases. e) Technology platforms must rapidly and correctly adapt, provide widespread offerings and scale appropriately, in the context of changing situational contexts.

We present Contour-net, a bio-inspired model for tactile contour-tracing driven by an Hopf oscillator. By controlling the rhythmic movements of a simulated insect-like feeler, the model executes both wide searching and local sampling movements. Contour-tracing is achieved by means of contact-induced phase-forwarding of the oscillator. To classify the shape of an object, collected contact events can be directly fed into machine learning algorithms with minimal pre-processing (scaling). Three types of classifiers were evaluated, the best one being a Support Vector Machine. The likelihood of correct classification steadily increases with the number of collected contacts, enabling an incremental classification during sampling. Given a sufficiently large training data set, tactile shape recognition can be achieved in a position-, orientation- and size-invariant manner. The suitability for robotic applications is discussed.

International audience; no abstract