Showing papers presented at "International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management in 2019"

A supervised multi-class multi-label word embeddings approach for toxic comment classification

Abstract: Nowadays, communications made by using the modern Internet-based opportunities have revolutionized the way people exchange information, allowing real-time discussions among a huge number of users. However, the advantages offered by such powerful instruments of communication are sometimes jeopardized by the dangers related to personal attacks that lead many people to leave a discussion that they were participating. Such a problem is related to the so-called toxic comments, i.e., personal attacks, verbal bullying and, more generally, an aggressive way in which many people participate in a discussion, which brings some participants to abandon it. By exploiting the Apache Spark big data framework and several word embeddings, this paper presents an approach able to operate a multi-class multi-label classification of a discussion within a range of six classes of toxicity. We evaluate such an approach by classifying a dataset of comments taken from the Wikipedia’s talk page, according to a Kaggle challenge. The experimental results prove that, through the adoption of different sets of word embeddings, our supervised approach outperforms the state-of-the-art that operate by exploiting the canonical bag-of-word model. In addition, the adoption of a word embeddings defined in a similar scenario (i.e., discussions related to e-learning videos), proves that it is possible to improve the performance with respect to solutions employing state-of-the-art word embeddings.

Prediction of Public Procurement Corruption Indices using Machine Learning Methods

Abstract: The protection of citizens’ public financial resources through advanced corruption detection models in public procurement has become an almost inevitable topic and the subject of numerous studies. Since it almost always focuses on the prediction of corrupt competition, the calculation of various indices and indications of corruption to the data itself are very difficult to come by. These data sets usually have very few observations, especially accurately labelled ones. The prevention or detection of compromised public procurement processes is definitely a crucial step, related to the initial phase of public procurement, i.e., the phase of publication of the notice. The aim of this paper is to compare prediction models using text-mining techniques and machine-learning methods to detect suspicious tenders, and to develop a model to detect suspicious one-bid tenders. Consequently, we have analyzed tender documentation for particular tenders, extracted the content of interest about the levels of all bids and grouped it by procurement lots using machine-learning methods. A model that includes the aforementioned components uses the most common text classification algorithms for the purpose of prediction: naive Bayes, logistic regression and support vector machines. The results of the research showed that knowledge in the tender documentation can be used for detection suspicious tenders.

Tacit Knowledge Transfer: Information Technology Usage in Universities

Abstract: The importance of transferring tacit knowledge transfer is acknowledged in the literature, but the usage of information technology for tacit knowledge transfer is not well researched. Through a mixed methods approach, employing an online questionnaire and interviews, this study explored the perceptions of university academic staff with regards to information technology usage, specifically relating to the transfer of tacit knowledge. The study found a lack of specific tacit knowledge transfer technologies but relatively high use of communication tools, a need for training on the use of new information technology was identified and academic staff are generally quick to adapt to information technology. However, there appeared to be a lack of confidence in information technology for the transfer of tacit knowledge and staff willingness to use technology for sharing tacit knowledge was not high, exhibiting uncertainty. This study contributes to a better understanding of the usage of information technology for tacit knowledge transfer and its adaptability by university academics. The results of this study may stimulate future research by addressing sample size limitation and replication in a different organisational setting.

An Overview of Blockchain for Higher Education

Abstract: Blockchain technology is one of the most widely acclaimed technologies of recent years. It enables the creation of distributed applications involving multiple actors or organizations in which transactions and data are not under the control of a central authority. Transactions are stored in a distributed public ledger in an immutable format in such a way that they can be verified by participants. This disintermediation promises to remove much of the “friction” (and lower costs) from distributed transactions by cutting out the intermediary party, thus enabling a variety of applications in finance, government, health, etc. This paper provides an overview of recent research in the application of blockchain technology to higher education for those who are interested in working in this area or just understanding how higher education may be positively impacted in the future by blockchain.

The Overview of Digital Twins in Industry 4.0: Managing the Whole Ecosystem

Abstract: Industry 4.0 aims in renewing processes using available technologies such as robots and other AI techniques implemented in IoT, drones, digital twins and clouds. This metamorphose impacts the whole industry ecosystems including people, information processing and business models. In this context, the accumulated knowledge and know-how can be reused but has also to evolve. This paper focus on the role of digital twins in transforming industrial ecosystems and discuss also the environmental impact.

D7-R4: software development life-cycle for intelligent vision systems

Abstract: Intelligent Vision Systems (IVS) are omnipresent in our daily life from social media apps to m-health services, from street surveillance cameras to airport e-gates, from drones to companion robots. Hence, IVS encompass any software which has a visual input processed by means of algorithm(s) involving Artificial Intelligence (AI) methods. The design and development of these IVS softwares has become an increasingly complex task, since vision-based systems have evolved into (semi-)autonomous AI systems, usually requiring effective and ethical data processing along with efficient signal processing and real-time hardware/software integration as well as User Experience (UX) and (cyber)security features. Consequently, IVS system development necessitates an adapted software development life-cycle (SDLC) addressing these multi-domain needs, whilst being developer friendly. Hence, we propose in this paper a new SDLC we called D7-R4 which allows developers to produce quality, new-generation IVS to be deployed in real-time and in real-world, unstructured environments.

A Framework for the Assessment and Training of Software Refactoring Competences

Abstract: Long-living software systems are becoming increasingly complex and difficult to maintain. Software refactoring is considered important to achieve maintainability and extensibility of a software system over time. In practice, it is still often neglected, partly because of costs, the perceived risks of collateral damage and difficulties of individuals working on certain components of complex software. It is therefore important for software projects that software developers have the appropriate skills and competences to efficiently perform software refactoring. However, so far there is no systematization of competences in software refactoring to guide in the assessment or training of competences, e.g., for planning or evaluating training activities and paths. In this paper, we address this need by presenting a competence framework for software refactoring by applying Bloom’s revised taxonomy for educational objectives. In particular, we specify competence levels by combining knowledge and cognitive-process dimensions. Via a case study with two existing training environments (i.e. a tutoring system and a serious game), we demonstrate by example that the framework can support (1) in analyzing the competence levels addressed by the training environments and (2) in reflecting training paths for software refactoring. Finally, we discuss the limitations and the further potential of the framework.

A Discretized Extended Feature Space (DEFS) Model to Improve the Anomaly Detection Performance in Network Intrusion Detection Systems

Abstract: The unbreakable bond that exists today between devices and network connections makes the security of the latter a crucial element for our society. For this reason, in recent decades we have witnessed an exponential growth in research efforts aimed at identifying increasingly efficient techniques able to tackle this type of problem, such as the Intrusion Detection System (IDS). If on the one hand an IDS plays a key role, since it is designed to classify the network events as normal or intrusion, on the other hand it has to face several well-known problems that reduce its effectiveness. The most important of them is the high number of false positives related to its inability to detect event patterns not occurred in the past (i.e. zero-day attacks). This paper introduces a novel Discretized Extended Feature Space (DEFS) model that presents a twofold advantage: first, through a discretization process it reduces the event patterns by grouping those similar in terms of feature values, reducing the issues related to the classification of unknown events; second, it balances such a discretization by extending the event patterns with a series of meta-information able to well characterize them. The approach has been evaluated by using a real-world dataset (NSL-KDD) and by adopting both the in-sample/out-of-sample and time series cross-validation strategies in order to avoid that the evaluation is biased by over-fitting. The experimental results show how the proposed DEFS model is able to improve the classification performance in the most challenging scenarios (unbalanced samples), with regard to the canonical state-of-the-art solutions.

How Are Situation Picture, Situation Awareness, and Situation Understanding Discussed in Recent Scholarly Literature?

Abstract: There are several different definitions of situation awareness. However, all of them have in common is knowing and understanding of what is happening, an understanding of future changes or problems, and the prediction of the future situation and the decisions to be made on its basis. Situation picture and Situation Awareness are narrow. Situation understanding of the situation is the understanding of the decision-makers and their assistants about what has happened, the circumstances that have affected them, the goals of the different parties and the possible development options of the events needed to make decisions on a particular issue or subject. The results of this study indicate that the recent discussion in scholarly literature focus on situation awareness. A further result is that the context of the many of the recent literature are focused on issues related to cyber security or on intelligent systems, thus on IT systems, which are very relevant to modern situation awareness and understanding in these modern times where more and more systems become digitalized and interconnected..

Complex Authority Network Interactions in the Common Information Sharing Environment

Abstract: European authorities collaborate as a community toward a coherent approach of situational understanding and open trust base information sharing. Innovation in multi-stakeholder collaboration networks involve complex collaboration between user community members, providing cross-sector, cross-border and cross-authority interaction and information sharing for collaborative situation awareness, and cooperation to increase safety and security. This study analyses data consisting of elements of use cases, collected from EU funded innovation projects. These were placed in a table based on similarity, difference and relevance to produce a classification. The results of this study indicate that use cases and scenarios engage end-users to co-create very practical descriptions providing input communication for innovation projects; also multi-actor projects are complex networks thus, this study contributes to the network approach of innovation. The implications of this study are that reaching faster innovation can be facilitated by leading and organising projects well, providing appropriate feedback to ensure project plans and results stay connected with project goals, fostering project continuums, and having e.g. higher education institutions bring problems as project ideas. The results, innovations, and feedback from research and innovation projects can benefit the European society.

Evaluation of the Performance of Information Systems Implemented at the University of Tripoli, Libya

Abstract: Evaluating the performance of information systems (ISs) has emerged from the increasing influence of information technology on the effectiveness and efficiency of work processes in an organization (Bryman and Bell 2007). The aim of the overall study is to overcome a lack in the literature regarding the assessment of information systems (IS) in Libyan Higher Education (LHE), especially universities. The aim of this initial article is to focus on the University of Tripoli (UOT), a study that will be extended to other Libyan public universities. A description of the study, its significance and objectives and the methodology followed are presented, together with an analysis of the findings on the basis of appropriately chosen models. Finally, we assess the current level of ISs implemented in UOT by analyzing the findings based on these models.

Blockchain or Distributed Ledger Technology What Is in It for the Healthcare Industry

Abstract: Distributed ledger technology has seen its debut into communities of practice in healthcare where the reliance on knowledge sharing between participants postulates the foundations of secure and distributed knowledge, especially in some sensitive context, such as patient information. This knowledge is essential for the practice of care from patient contact to research, pharmaceutical supply chain, medication adherence and management of the plethora of bedside data into a collection of knowledge about the patient, essential to quality care. We introduce different schools of thought and implementation contexts of the distributed ledger technology or Blockchain. We provide an overview of Blockchain and Distributed Ledger Technology, focused on the Healthcare industry, as an initial assessment of the validity of an application of Distributed Ledger Technology in a specific knowledge management model to solve problems related to knowledge sharing in medical knowledge management systems. The paper summarizes some instances of most likely and unlikely uses of Blockchain in the healthcare setting. The paper also introduces a few use cases where some short-term benefits from such implementation.

Collaborative Knowledge Management in University Alliances with Information Models

Abstract: Alliances between enterprises, such as Star Alliance, are a well-known phenomenon and have been subject of research for the last decades. Today, universities are also beginning to form alliances among themselves. Especially in the area of knowledge transfer alliances matter, as they create synergies, increase the visibility and allow universities to carry out projects that cannot be done by a single university. However, a University alliance creates new processes and interfaces between the member Universities. The management of such an alliance is a knowledge management challenge on its own. Therefore, this paper gathers the requirements on a University alliance and outlines how the business processes, that are specific for a University alliance, can be structured in a framework. The framework indicates which processes are important for an alliance and on which level they have to be addressed, on the level of a single University, first at each University and afterwards in the alliance or on alliance level only.

InnoDeck: Card based Innovation Support - A Modular Human-Centered Approach to Facilitate Innovation Workshops

Abstract: Innovation drives economic growth and today innovation workshops are widely used to create new products, production methods etc. The InnoDeck is a low-tech knowledge management concept for information that is relevant for the facilitation of innovation workshops. In this context, InnoDeck is a tool for information sharing and organizational learning. It consists of two-sided self-contained cards that provide either methodological or inspirational content. Facilitators can choose a subset of cards for their design thinking project. The InnoDeck is human-centered because the main focus of each card is to be a quick read, easy to grasp and memorable. It not only engages its users but also is highly participatory. Everyone is encouraged to add cards to the expandable accumulative card deck. The concept has been developed, used and evaluated within a network of insurance companies and has proven to be beneficial for creating an innovation culture within these companies.

Rethinking Strategies of Hackathons to Increase Team’s Creativity: Findings of a Qualitative Research

Abstract: Hackathons are events that have become increasingly common around the world. This kind of event, described as a programming marathon, is based on problem-solving that can go beyond the technological boundary. This paper presents the findings of an international hackathon to aid its organizers to rethink their strategies to improve the development of the team’s creativity to solve the challenge proposed. The paper summarizes qualitative research based on interviews and observations which point out that the organizers should consider strategies to improve knowledge application, resolving conflicts, individual learning, and experienced emotions, during pre-hackathon as well as post-hackathon events. Our findings could leverage the innovation, creativity, and knowledge sharing and creation within hackathons.

Information Modeling of Rule-based Logistic Planning Processes Kanban Loop Planning Supported by a Workflow Engine

Abstract: This paper discusses the modeling of rule-based logistics planning processes. These are mostly inadequately documented and modeled, especially for small and medium-sized enterprises (SMEs). As a starting point, the ways of representing rule-based logistics planning processes and the modeling languages suitable for the processes are introduced. In addition, it is shown how decision rules can be represented in modeling languages. Based on this, a prototypical representation for the planning of a kanban loop is presented as a technical model. This serves as the basis for a workflow, which is constructed by transforming the domain-oriented model into a technical model. A workflow engine is used to execute and evaluate the technical model.

What Promotes Intention? Factors Influencing Consumers’ Intention to Purchase Animal-Welfare Friendly Beef in Japan

Abstract: Food industries are required to face both increasing demand from a growing population with social development and enhancement of its sustainability. Farm animal welfare has become an important aspect of sustainable business development, but is still an unfamiliar concept for consumers in Japan, although Japanese society is under pressure to catch up with global trends. Researchers have been working around the world to explore consumer behavior concerning animal welfare in markets, but few such studies have been performed in Japan. In this study, we explored factors influencing consumers’ intention to purchase animal welfare friendly beef products (AWFBP) in Japan, by considering them as food products purchased involving ethical decision-making and empathy for beef cattle and farmers. An online questionnaire was used to identify consumer characteristics and perceived attributes of AWFBP among 620 consumers in the three largest cities in Japan. Based on the Theory of Planned Behavior, we found that perception of attributes perceived behavioral controls, and empathy for beef cattle were likely to influence consumers’ intention to purchase AWFBP.

Knowledge Management Models for the Smart Factory: A Comparative Analysis of Current Approaches

Abstract: This paper analyses current and well-known knowledge management models regarding their applicability to smart factories and the Industry 4.0. In form of a literature study, we surveyed the specific challenges and requirements that smart factories and the ongoing digital transition in the industrial sector introduce to knowledge management systems and models. In the second part, we then expound the extent to which those requirements are supported by well-established knowledge management models in form of a comparative analysis. A central result of this work is that an Industry 4.0 compliant knowledge management needs to incorporate aspects that emphasize human-machine and machine-machine interaction together with data protection and privacy concerns, besides other well-researched and established aspects.

Quality of Wikipedia Articles: Analyzing Features and Building a Ground Truth for Supervised Classification.

Abstract: Wikipedia is nowadays one of the biggest online resources on which users rely as a source of information. The amount of collaboratively generated content that is sent to the online encyclopedia every day can let to the possible creation of low-quality articles (and, consequently, misinformation) if not properly monitored and revised. For this reason, in this paper, the problem of automatically assessing the quality of Wikipedia articles is considered. In particular, the focus is (i) on the analysis of groups of hand-crafted features that can be employed by supervised machine learning techniques to classify Wikipedia articles on qualitative bases, and (ii) on the analysis of some issues behind the construction of a suitable ground truth. Evaluations are performed, on the analyzed features and on a specifically built labeled dataset, by implementing different supervised classifiers based on distinct machine learning algorithms, which produced promising results.

Towards Improving Knowledge Capitalization System for Sport Events Legacy.

Abstract: Knowledge Management is a way to answer the problem of capitalizing on the companies’ knowledge. Knowing that hosting sports events (SE) requires organizers to learn from past events to not repeat mistakes, we examine knowledge management in a sport events legacy (SEL). Thus, in this paper, we propose in first, two conceptual models based on UML; one for the SE, another for SEL. Secondly, we propose a system to manage SEL to assist in the process of data acquisition and capitalization on SE knowledge. This system helps to create an open collaborative platform for consultation, visualization of the spinoffs of sport events. It is intended to be used by public policies, territories, journalists, citizens, historians and all others. We propose also to take into account the spatiotemporal aspects of SE.

Formalizing Graphical Modularization Approaches for Ontologies and the Knowledge Loss

Abstract: This paper formalizes the graphical modularization technique view traversal for an ontology component of a Domain Information System (DIS). Our work is motivated by developing the ability to dynamically extract a module (called a view traversal module) based on an initial set of concepts. We show how the ability to quantify the knowledge that is preserved (or lost) in a view traversal module is significant for a multi-agent setting, which is a setting that requires provable privacy. To ensure partial knowledge preservation, we extend the view traversal module to a principal ideal subalgebra module. The cost of this extension is that the obtained knowledge is coarser, as the atoms of the associated lattice are composite yet considered atomic. The presented work constitutes a foundational step towards theories related to reasoning on partial domain knowledge.

Blockchain Certification and Granular Editing Permissions in Document Management System

Abstract: Ever-growing digitalization and increasingly competitive markets are driving industry and the public sector into fast-paced transformation. Competitive advantage is being acquired through technology investments made possible by previously unavailable resources, freed by process automation, simplification, and rationalization. Under these contingencies, we propose an innovative document management platform, featuring a collaborative document editing technique and a blockchain certification procedure. The two proposing parties - a private company and an academic organization - mutually agreed on employing open-source technologies as a strategic means to promote software reuse and developer communities’ support, and consequently reduce implementation costs and ensure interoperability.

Enterprise Transformation Management based on Enterprise Engineering Approach with Unified Enterprise Transformation Dimensions

Abstract: In the enterprise transformation (ET), there are so many ideal models, blueprints and situations. The ideal pictures are provided by practitioners and researchers one picture by one change is predicted or occurs on the business environment, for example, “digital enterprise transformation” by “business model at digital age”, etc. Indeed, a variety of approaches were proposed in the literature. On the other hand, under our literature survey, existing management frameworks are addressing one specific perspective of enterprise management and focusing on one kind of measurement. There is no significant adoption in the state of the enterprise transformation management systems based on the relationship between architecture and transformation practices yet. The goal of this work is, therefore, to propose a holistic management framework to support the transformation based on enterprise engineering. All the dimensions, analysis perspectives, impact analysis of those change practices together support among adaptable enterprise architecture world and real transformation world. It aims to enable the framework to be used in state-of-the-art enterprise change environments.

Knowledge-based Service for African Traditional Herbal Medicine: A hybrid approach

Abstract: Globally, the acceptance and use of herbal and traditional medicine is on the rise. Africa, especially Ghana, has its populace resorting to African Traditional Herbal Medicine (ATHMed) for their healthcare needs due to its potency and accessibility. However, the practice involving its preparation and administration has come into question. Even more daunting is the poor and inadequate documentation covering the preservation and retrieval of knowledge on ATHMed for long-term use, resulting in invaluable healthcare knowledge being lost. Consequently, there is the need to adopt strategies to help curtail the loss of such healthcare knowledge, for the benefit of ATHMed stakeholders in healthcare delivery, industry and academia. This paper proposes a hybrid-based computational knowledge framework for the preservation and retrieval of traditional herbal medicine. By the hybrid approach, the framework proposes the use of machine learning and ontology-based techniques. While reviewing literature to reflect the existing challenges, this paper discusses current technologies suited to approach them. This results in a framework that embodies an ontology driven knowledge-based system operating on a semantically annotated corpus that delivers a contextual search pattern, geared towards a formalized, explicit preservation and retrieval mechanism for safeguarding ATHMed knowledge.

Assist the Sustainable Development within Industries through the Territorial Knowledge Ontology

Abstract: Studying of territory as the main dimension of sustainability impact in the industrial activities and decision makers’ information when considering the sustainability in their activities. Therefore, exploring of territorial knowledge in order to integrate into industries activates is needed. So, this research is proposed a descriptive ontology for territorial knowledge (DOTK) which make explicit the knowledge of actors within industries about the sustainable development goal. Also, implementation of this ontology to a real case is proved that it can identify the intangible and tangible resources of territory for sustainability. Moreover, a semantic graph is proposed which shows the relationships between entities of DOTK ontology. Final validation of DOTK ontology is performed by the interview with the organizations of sustainable development implementation.

Action Representation for Intelligent Agents Using Memory Nets

Abstract: Memory Nets (Eggert et al.: Memory Nets: Knowledge Representation for Intelligent Agent Operations in Real World, 2019) are a knowledge representation schema targeted at autonomous Intelligent Agents (IAs) operating in real world. Memory Nets are targeted at leveraging the large body of openly available semantic information, and incrementally accumulating additional knowledge from situated interaction. Here we extend the Memory Net concepts by action representation. In the first part of this paper, we recap the basic domain independent features of Memory Nets and the relation to measurements and actuator capabilities as available by autonomous entities. In the second part we show how the action representation can be created using the concepts of Memory Nets and relate actions that are executable by an IA with tools, objects and the actor itself. Further, we show how action specific information can be extracted and inferred from the created graph. The combination of the two main parts provide an important step towards a knowledge base framework for researching how to create IAs that continuously expand their knowledge about the world.

Knowledge Requirements for Sustainable Smart Service Design.

Abstract: Recent research have addressed the topic of smart services from distinct angles, covering both technical and business aspects. However, a holistic approach in development processes of such services have yet to be fully covered. Therefore, this paper proposes an elicitation of requirements process as the initial step of a smart service design approach. The process takes information and knowledge needs as its core element for development, also considering customer centricity, service lifecycle, and sustainability concerns. A text mining tool was used to discover the unknown knowledge requirements from different text-data sources presented in a case ecosystem. After a co-occurrence analysis performed by our text mining software, we extracted the most relevant natural linguistic elements, which are expressed as knowledge requirements. The proposed elicitation process aims to lay the foundations for further propositions with a holistic point of view. Future research could aim the application of other technologies and methods for service design, as well as a broader approach in business processes and interdisciplinary cooperation.

Knowledge Management in Healthcare: Information Requirements When Creating a Decision Support Tool in Radiology

Abstract: Introduction: This empirical work examines the information requirements when undertaking a process modelling project in a Healthcare setting such as a CT (Computed Tomography) department. Using qualitative and quantitative methods we map the process, incorporating patient, staff and process related components so as to quantify resource utilisation and the service experienced by the patient. Method: In this study, semi structured interviews are used to identify patient complexity factors/characteristics. Process mapping and involvement of stakeholders are discussed as is the identification and analysis of data. A discrete event simulation (DES) model of the process is designed and performance metrics identified. Results: Yearly demand for Radiology services are increasing significantly. Factors determining patient complexity and variation include patient type, infectiousness, mobility, exam type and patient care needs. A strong correlation between age and infectiousness was observed. Conclusion: DES modelling, though data intensive, provides decision makers with insights into resource utilisation, process capacity, delays and disruptions and in doing so supports operations, management and the adoption of good practices in Healthcare.

Topological Approach for Finding Nearest Neighbor Sequence in Time Series

Abstract: The aim of this work is to obtain a good quality approximation of the nearest neighbor distance (nnd) profile among sequences of a time series. The knowledge of the nearest neighbor distance of all the sequences provides useful information regarding, for example, anomalies and clusters of a time series, however the complexity of this task grows quadratically with the number of sequences, thus limiting its possible application. We propose here an approximate method which allows one to obtain good quality nnd profiles faster (1-2 orders of magnitude) than the brute force approach and which exploits the interdependence of three different topologies of a time series, one induced by the SAX clustering procedure, one induced by the position in time of each sequence and one by the Euclidean distance. The quality of the approximation has been evaluated with real life time series, where more than 98% of the nnd values obtained with our approach are exact and the average relative error for the approximated ones is usually below 10%.

A Two-Step Feature Space Transforming Method to Improve Credit Scoring Performance

Abstract: The increasing amount of credit offered by financial institutions has required intelligent and efficient methodologies of credit scoring. Therefore, the use of different machine learning solutions to that task has been growing during the past recent years. Such procedures have been used in order to identify customers who are reliable or unreliable, with the intention to counterbalance financial losses due to loans offered to wrong customer profiles. Notwithstanding, such an application of machine learning suffers with several limitations when put into practice, such as unbalanced datasets and, specially, the absence of sufficient information from the features that can be useful to discriminate reliable and unreliable loans. To overcome such drawbacks, we propose in this work a Two-Step Feature Space Transforming approach, which operates by evolving feature information in a twofold operation: (i) data enhancement; and (ii) data discretization. In the first step, additional meta-features are used in order to improve data discrimination. In the second step, the goal is to reduce the diversity of features. Experiments results performed in real-world datasets with different levels of unbalancing show that such a step can improve, in a consistent way, the performance of the best machine learning algorithm for such a task. With such results we aim to open new perspectives for novel efficient credit scoring systems.