DOI: 10.3897/jucs.152864

Authors: Claudio Alvarez, Andres Carvallo, Gustavo Zurita

Abstract: As student cohorts grow, real-time case-based learning discussions generate increasing volumes of textual data, intensifying the orchestration load teachers must manage. Reviewing and providing feedback on student responses promptly becomes increasingly challenging, demanding efficient methods to assist educators in selecting relevant contributions to steer classroom discussions. This study proposes a low-footprint natural language processing (NLP) approach that leverages small-scale models running on commodity hardware, avoiding the computational overhead and cost associated with large language models. Our system, integrated into EthicApp, a collaborative learning platform, employs pre-trained language models such as the Universal Sentence Encoder (USE) and Bidirectional Encoder Representations from Transformers for Spanish (BETO), along with traditional text-mining techniques like Term Frequency-Inverse Document Frequency (TF-IDF). Through expert evaluations, we found that BETO exhibited superior performance in identifying relevant student responses but required GPU acceleration. At the same time, USE provided an efficient alternative that outperformed TF-IDF and remained feasible for CPU-based execution. Additionally, the methods showed a tendency—most notably BETO—to select longer responses, which, rather than introducing selection bias, was interpreted as an indicator of deeper student engagement. No significant semantic bias was found, ensuring a fair representation of students’ perspectives. Our findings suggest that low-footprint NLP can effectively reduce teacher orchestration load, enabling more targeted feedback without requiring extensive computational resources.

HTML

PDF

DOI: 10.3217/jucs-022-08-1048

Authors: Haroldo Maria Teixeira Filho, Leonardo Azevedo, Sean Wolfgand Matsui Siqueira

Abstract: Organizations are adopting Service-Oriented Architecture (SOA) to increase operation's efficiency and flexibility. To accomplish these goals, it is necessary to ensure that the architecture and its evolution are compliant with business goals, best practices, legal and regulatory requirements. However, the diversity of domains and stakeholders involved in SOA solutions demands complex and expensive work to validate the architecture compliance. Hence, it can result in high costs and low quality assessment if the organization does not use an effective approach in this scenario. In addition, it would be important to consider standards and open solutions in order to promote interoperability and reuse of available tools, making it easier to spread throughout the organizations. We propose intelliGov, an architecture that aims to solve these problems by using ontologies, semantic rules and queries in order to simplify the compliance validation process. The architecture employs open standards - OWL, SWRL and SQWRL - in its implementation. A case study, executed in a global energy company that is currently adopting SOA, demonstrates gains in quality and costs of the compliance assessment process using the proposed architecture.

HTML

XML

PDF

DOI: 10.3217/jucs-022-05-0709

Authors: Evangelos Kalampokis, Areti Karamanou, Efthimios Tambouris, Konstantinos Tarabanis

Abstract: Billions of people everyday use Social Media (SM), such as Facebook and Twitter, to express their opinions and experiences with brands. Companies are highly interested in understanding such SM brand-related content. Consequently, many studies have been conducted and many applications have been developed to analyse this content. For analysis purposes, the main SM metrics used include volume and sentiment. Interestingly, however, brand equity theory proposes different metrics for assessing brand reputation. These include brand image, brand satisfaction and purchase intention (henceforth referred to as marketing metrics). The objective of this paper is to explore the feasibility of applying marketing metrics in Twitter brand-related content. For this purpose, we collect, study and analyse tweets that mention two brands, namely IKEA and Gatorade. The manual analysis suggests that a significant amount of brand tweets is related to brand image, brand satisfaction and purchase intention. We thereafter design an algorithm that classifies tweets into relevant categories to enable automatic marketing metrics computation. We implement the algorithm using statistical learning approaches and prove that its classification accuracy is good. We anticipate that this article will motivate other studies as well as applications' designers in adopting marketing theories when evaluating brand reputation through SM content.

HTML

XML

PDF

DOI: 10.3217/jucs-020-10-1443

Authors: María García-Álvarez, Laura Varela-Candamio, Isabel Novo-Corti

Abstract: The current higher education programs use information and communication technologies to conduct interactive teaching and learning activities. This paper creates an educational method based on an Interaction Analysis Model through the use of chats in higher education. Compared to the traditional functions of the chats in education, our proposal introduces discussions of current economic events and real cases. This contributes to develop the problem-based learning and leads to students not only to improve their knowledge but to develop skills such as teamwork or leadership, which should be important characteristics of a graduate in Business Degree. As a result, students transfer their knowledge to solve current case studies improving their interest in the subject greatly and, therefore, their motivation and the social knowledge construction of the whole group.

HTML

XML

PDF

DOI: 10.3217/jucs-019-12-1808

Authors: Josiane Michalak Hauagge Dall Agnol, Cesar Tacla

Abstract: This paper proposes a framework of the negotiation process for solving divergences in the collaborative ontology development. Such framework is obtained through the use of philosophical principles deriving from the theories of essence, identity, unity and dependence (preconized by the OntoClean methodology) as to justify part of the argumentation used in the negotiation process among the participants, besides helping reach a consensus and reduce the conceptual gap among models. The evaluation of the experiments conducted with the use of the proposed method suggests the feasibility and implementability of our approach in practice.

HTML

XML

PDF

DOI: 10.3217/jucs-019-11-1543

Authors: Stefan Dietze, Eleni Kaldoudi, Nikolas Dovrolis, Daniela Giordano, Concetto Spampinato, Maurice Hendrix, Aristidis Protopsaltis, Davide Taibi, Hong Yu

Abstract: Research in technology-enhanced learning (TEL) throughout the last decade has largely focused on sharing and reusing educational resources and data. This effort has led to a fragmented landscape of competing metadata schemas, such as IEEE LOM or ADL SCORM, and interface mechanisms, such as OAI-PMH, SQI and REST-ful services in general. More recently, semantic technologies were taken into account to improve interoperability. However, so far Web-scale integration of resources is not facilitated, mainly due to the lack of take-up of shared principles, datasets and schemas. On the other hand, the Linked Data approach has emerged as the de facto standard for sharing data on the Web and is fundamentally based on established W3C standards. This paper presents results of the European Commission-funded project mEducator, which exploits Linked Data principles for (1) semantic integration and (2) social interconnecting of educational data, resources and actors. We describe a general approach to exploit the wealth of already existing educational data on the Web by allowing its exposure as Linked Data and by taking into account automated enrichment and interlinking techniques to provide a rich and well-interlinked graph for the educational domain. Additionally, the paper presents an evaluation of our work with respect to a set of socio-semantic dimensions. Experimental results demonstrate improved interoperability and retrievability of the resulting resource descriptions as part of an interlinked resource graph.

HTML

XML

PDF

DOI: 10.3217/jucs-018-19-2706

Authors: Anuradha Mathrani, David Parsons, Sanjay Mathrani

Abstract: Offshore software development (OSD) is a leading business sector in the global IT marketplace, and vendors in different countries are opening software development centres to take advantage of new business opportunities. However, software development is both a technical and a social process in which various software modules are integrated, requiring ongoing interaction and synchronisation of activities between distributed stakeholders. Knowledge management (KM) strategies are applied to create knowledge consistent with client requirements, project specific features and chosen design methodologies. Building on existing KM theories with empirical evidence from ten case studies in the Asia Pacific region, within two country contexts (New Zealand and India), this research reveals the KM initiatives for enabling knowledge transfer in the OSD process at the operational, design and strategic level. The paper offers insights on how software vendors build organisational knowledge repositories as they streamline distributed tasks in different country contexts. Country-specific contexts reveal that New Zealand vendors are engaged more in project and product management and have further outsourced software development tasks to other low cost countries. The Indian vendors are involved in software construction, development of technical specialist skills and use of more formal processes. These findings emphasise implications of various sociological, cultural and technical perspectives of KM initiatives in OSD.

HTML

XML

PDF

DOI: 10.3217/jucs-017-10-1455

Authors: Nicole Bischof, Martin Eppler

Abstract: Knowledge communication is an essential mechanism to facilitate intra- and inter-organizational knowledge transfer. In order to improve the efficiency of knowledge communication, organizations need to pay particular attention to the clarity of conveyed knowledge in order not to create confusion, misunderstandings, or misapplication of knowledge. In this contribution, we show where and how the concept of clarity matters for knowledge management in general, and for knowledge communication in particular. We review and operationalize the clarity concept so that it can become the object of a systematic management effort. Furthermore, we show ways of how clarity can be pro-actively and systematically managed. We have tested our conception of clarity in a survey on clarity in knowledge-focused presentations, and we present the results in this article. An outlook on future research on clarity in knowledge management concludes the contribution.

HTML

XML

PDF

DOI: 10.3217/jucs-017-10-1390

Authors: Ali Abdoli, Jamal Shahrabi, Jalil Heidary

Abstract: By entering the knowledge age and the appearance of knowledge economy, organizations are more dependent on knowledge workers productivity. Productivity means doing the right things right. It shows how a knowledge worker makes use of resources to fulfill the goals of the organization. This definition makes productivity be the result of simultaneous existence of efficiency "doing the things right" and effectiveness "doing the right things". Since factors influencing knowledge workers productivity cannot be definitely measured, uncertainty theory plays an important role in this area. So in this paper, first, dimensions of productivity will be introduced and then, by the use of linguistic fuzzy approach and DEA, efficiency and effectiveness of knowledge workers will be measured. Next, a model for measuring knowledge workers productivity will be presented on the basis of efficiency and effectiveness. Finally, values of knowledge workers productivities will be ranked. In the last section, the result of this five-step method is examined through a case study.

HTML

XML

PDF

DOI: 10.3217/jucs-017-02-0203

Authors: Naoufel Cheikhrouhou, Michel Pouly, Charles Huber, Alok Choudhary

Abstract: Small and Medium Enterprises (SMEs) face new challenges in the global market as customers require more complete and flexible solutions and continue to drastically reduce the number of suppliers. SMEs are trying to address these challenges through cooperation within collaborative enterprise networks (CENs). Human aspects constitute a fundamental issue in these networks as people, as opposed to organizations or Information Technology (IT) systems, cooperate. Since there is a lack of empirical studies on the role of human factors in IT-supported collaborative enterprise networks, this paper addresses the major human aspects encountered in this type of organization. These human aspects include trust issues, knowledge and know-how sharing, coordination and planning activities, and communication and mutual understanding, as well as their influence on the business processes of CENs supported by IT tools. This paper empirically proves that these aspects constitute key factors for the success or the failure of CENs. Two case studies performed on two different CENs in Switzerland are presented and the roles of human factors are identified with respect to the IT support systems. Results show that specific human factors, namely trust and communication and mutual understanding have to be well addressed in order to design and develop adequate software solutions for CENs.

HTML

XML

PDF

DOI: 10.3217/jucs-013-10-1471

Authors: Chunguo Wu, Maurizio Marchese, Jingqing Jiang, Alexander Ivanyukovich, Yanchun Liang

Abstract: With the currently growing interest in the Semantic Web, keywords/metadata extraction is coming to play an increasingly important role. Keywords extraction from documents is a complex task in natural languages processing. Ideally this task concerns sophisticated semantic analysis. However, the complexity of the problem makes current semantic analysis techniques insufficient. Machine learning methods can support the initial phases of keywords extraction and can thus improve the input to further semantic analysis phases. In this paper we propose a machine learning-based keywords extraction for given documents domain, namely scientific literature. More specifically, the least square support vector machine is used as a machine learning method. The proposed method takes the advantages of machine learning techniques and moves the complexity of the task to the process of learning from appropriate samples obtained within a domain. Preliminary experiments show that the proposed method is capable to extract keywords from the domain of scientific literature with promising results.

HTML

XML

PDF

DOI: 10.3217/jucs-013-10-1411

Authors: Spyros Sioutas, Kostas Tsichlas, Bill Vassiliadis, Dimitrios Tsolis

Abstract: Indexing video content is one of the most important problems in video databases. In this paper we present linear time and space algorithms for handling video metadata that represent objects or events present in various frames of the video sequence. To accomplish this, we make a straightforward reduction of this problem to the intersection problem in Computational Geometry. Our first result is an improvement over the one of V. S. Subrahmanian [Subramanian, 1998] by a logarithmic factor in storage. This is achieved by using different basic data structures. Then, we present two other interesting time-efficient approaches. Finally a reduction to a special geometric problem is considered according to which we can achieve two optimal in time and space solutions in main and external memory model of computation respectively. We also present an extended experimental evaluation.

HTML

XML

PDF

DOI: 10.3217/jucs-013-09-1287

Authors: José Troyano, Fernando Enríquez, Fermín Cruz, José Cañete-Valdeón, F. Ortega

Abstract: In this paper we present an experience in the extraction of named entities from Spanish texts using stacking. Named Entity Extraction (NEE) is a subtask of Information Extraction that involves the identification of groups of words that make up the name of an entity, and the classification of these names into a set of predefined categories. Our approach is corpus-based, we use a re-trainable tagger generator to obtain a named entity extractor from a set of tagged examples. The main contribution of our work is that we obtain the systems needed in a stacking scheme without making use of any additional training material or tagger generators. Instead of it, we have generated the variability needed in stacking by applying corpus transformation to the original training corpus. Once we have several versions of the training corpus we generate several extractors and combine them by means of a machine learning algorithm. Experiments show that the combination of corpus transformation and stacking improve the performance of the tagger generator in this kind of natural language processing applications. The best of our experiments achieves an improvement of more than six percentual points respect to the predefined baseline.

HTML

XML

PDF

DOI: 10.3217/jucs-013-09-1234

Authors: Muhammad Afzal, Narayanan Kulathuramaiyer, Hermann Maurer

Abstract: We are approaching an era where research materials will be stored more and more as digital resources on the World Wide Web. This of course will enable easier access to online publications. As the number of electronic publications expands, it will, however, become a challenge for individuals to find related or relevant papers. Related papers could be papers written by the same team of authors or by one of the authors, or even papers that deal with the same topic but were written by other authors. This, of course, raises the issue of linking to papers forward in time, or as we call it "links into the future". To be concrete, while reading a paper written in the year 1980, it would be nice to know if the same author has written another related paper in 1990’s or if the same author has written a paper earlier, all this without making an explicit search. Based on the ascertained interest of a person reading a particular paper from a digital repository, an auto-suggestion facility could be useful to indicate papers in the same area, category and subject that might potentially be of interest to the reader. One is typically interested in finding related papers by the same author or by one of the authors of a paper. This feature can be implemented in two ways. The first is by creating links from this paper to all the relevant papers and updating it periodically for new papers appearing on the World Wide Web. Another way is by going through the references of all papers appearing on the WWW. Based on the references, one can create mutual links to the papers that are referred to.

HTML

XML

PDF

DOI: 10.3217/jucs-013-09-1184

Authors: Simona Colucci, Tommaso Noia, Eugenio Di Sciascio, Francesco Donini, Azzurra Ragone

Abstract: Knowledge management is characterized by many different activities ranging from the elicitation of knowledge to its storing, sharing, maintenance, usage and creation. Skill management is one of such activities, with its own peculiarities, as it focuses on full exploitation of knowledge individuals in an organization have, in order to carry out at best given tasks. In this paper a semantic-based automated Skill Management System is proposed, which supports competences search and creation. The system implements an approach exploiting the formalism and the reasoning services provided by Description Logics. The approach embeds also non standard Description Logics reasoning services to extend the set of provided features. Here we present main characteristics of our system and focus on a novel algorithm exploiting advanced inference services for the one-to-one assignment of a set of individuals to a set of tasks, endowed of logical explanation features for missing/conflicting skills.

HTML

XML

PDF

DOI: 10.3217/jucs-013-09-1157

Authors: Sinuhé Arroyo

Abstract: Ontologies have been recognized as a fundamental infrastructure for advanced approaches to Knowledge Management (KM) automation in SOA. Building services communicate with each other by exchanging self-contained messages. Depending on the specific requirements of the business model they serve and the application domain for which services were deployed, a number of mismatches (i.e. sequence and cardinality of messages exchanges, structure and format of messages and content semantics), can occur which prevent interoperation among a prior compatible services. Existing choreography technologies attempt to model such external visible behavior. However, they lack the consistent semantic support required to fully meet the necessities of heterogeneous KM environments. This paper describes the ontology and grammar of SOPHIE, a semantic service-based choreography framework for overcoming conversational pattern mismatches in knowledge intensive environments. Consequently, the paper provides an overview of the framework that depicts its main building blocks, so a good understantind of the ontology and grammar that summarize the conceptual model is gained. Such ontology allows the desing and description of fully fledged choreographies that can be used, as a result of a mediation task, to produce the mediating structures that in fact allow dynamic service-to-service interoperation. Finally, a use case centred in the telcomunications field serves as proof of concept of how SOPHIE is being applied.

HTML

XML

PDF