ONTOLOGY-BASED LEARNING OBJECTS SEARCH AND COURSES GENERATION

Learning objects retrieval is important for a variety of information needs and applications such as collection management, summary and analysis. Especially, retrieval of learning objects from the big collection of textual database has become a very active field. Since the learning information is stored in textual format in most of the time, the retrieval of learning objects from textual database faces two major challenges such as, large data handling, and effectiveness. If these two challenges are solved, the performance and its applicability will be improved significantly. Accordingly, the first challenges of large data processing will be handled through the semantic annotations method. In the semantic annotation method, the text document was converted to semantic annotation data using concept-based modeling. With the help of existing work, the large data will be converted to annotation object so that the matching process with the input query will be reduced and the complexity of handling big data will be also reduced. The second challenge of effectiveness will be solved using new matching score to obtain better retrieval effectiveness. Here, query will be matched with the annotated object using matching score that will be devised newly and it will be applied for learning object retrieval. The experimentation will be carried out utilizing the learning objects given in IEEE digital library and the performance of the proposed technique is evaluated using precision, recall and F-measure and also, comparative analysis will be performed to prove the better performance of the proposed technique. The analysis from the experimentations showed that the proposed approach has obtained a maximum precision of 0.98 and average precision of 0.95.

This paper proposes a new approach to automatic retrieval and composition of Learning Objects (LOs) in an Adaptive Educational Hypermedia System (AEHS) using multidimensional learner characteristics to enhance learning effectiveness. The approach focuses on adaptive techniques in four components of AEHS: Learning Paths, LO Retrieval, LO Sequencing, and Examination Difficulty Levels. This approach has been designed to enable the adaptation of rules to become generic. Hence, the application to various domains is possible. The approach dynamically selects, sequences, and composes LOs into an individual learning package based on the use of domain ontology, learner profiles, and LO metadata. The Sharable Content Object Reference Model is employed to represent LO metadata and learning packages in order to support LO sharing. The IMS Learner Information Package Specification is used to represent learner profiles. A preliminary evaluation of the developed system indicates the system’s effectiveness in terms of learners’ satisfaction.

This paper proposes a multiagent system application model for indexing, retrieving and recommending learning objects stored in different and heterogeneous repositories. The objects within these repositories are described by filled fields using different metadata (data about data) standards. The searching mechanism covers several different learning object repositories and the same object can be described in these repositories by the use of different types of fields. Aiming to improve accuracy and coverage in terms of recovering a learning object and improve the relevance of the results we propose an information retrieval model based on a multiagent system approach and an ontological model to describe the covered knowledge domain.

[EN] The World Wide Web commonly known as Web as evolve and nowadays passes through a transition to a third generation, Web 3.0. In parallel, different areas of application as also evolve. One of them was eLearning, with the development of several systems and tools. However, despite many years of research and the appearance of the concept of Semantic Web for quite a time, the Web and consequently eLearning systems and mainly the user’s still seem not to be prepared for this transition. There are different reasons for this situation. First, the application of concept requires specialized technical knowledge of the users to design the systems and resources accordingly to the educational standards and semantic technologies. Consequently, their dissemination outside experimental contexts is limited. Moreover, since they lack application in eLearning systems of Semantic Web, and lack some basic knowledge regarding standardization of learning design, resources and also reusability and the stability of these concepts before go a step forward to semantic web. Thus, lacks for application systems that make the transition easier for the users. In order to solve these problems, this thesis proposes the development of a web-based information system – AHKME (Adaptive Hypermedia Knowledge Management Elearning System) - that combines concepts of Web 2.0 and 3.0, regarding advanced collaboration, social networks, interoperability, standardization and adaptation, in a concept that we prefer to call Web/eLearning 2.5. The objective of this system is to prepare the users for the new paradigm and the application of concepts regarding reusability, giving flexibility to the users by offering tools to facilitate the learning design and even the creation of their own personalized schemas/structure for instructional design. AHKME is based on schemas of standards and specifications (e.g. IMS) for standardization of resources, data mining and workflow for adaptation, and user’s interactivity with learning resources and collaboration through social networks and players. Morevoer, an evaluation of this proposal was performed through laboratory tests to evaluate the results and behaviour of AHKME, simulating a real learning situation. The main goal of this study is to open the way for new researchs and systems for preparing the users for this new generation of Web, and by this mean introduce the semantic technologies, cross-platforms systems and “intelligent web”.

In the nineties, two different trends on distance learning research arose. The first one aimed to create systems able to adapt themselves to students’ characteristics, based on instructional and learning theories, while the second aimed to facilitate the creation and management of distance courses, by reusing material. Technologies available so far had hindered the creation of systems able to conjoin the benefits achieved by both groups because adaptive systems were limited to a specific knowledge domain, and almost everything had to be recreated to be used in a different subject. On the other hand, systems able to deal with different subjects could not automatically adapt their content to students’ needs. The emergence of the Semantic Web, as well as Learning Objects, led researchers to develop projects based on them to solve that problem. At last, adaptive and reusable courses based on learning theories could be developed. Learning Object Metadata and other standards for e-learning, as well as ontologies about different knowledge domains, students’ characteristics and learning theories were the tools that researchers were missing to solve such problems. Although much research and publications have been done in that period, few practical results were noticed, frustrating not only the students, but also the teachers and other education-related workers, as they realized that the level of system autonomy they had aimed for had not been reached. This paper describes the euphoria caused by the Semantic Web, problems related to the engineering of ontologies that led to a frustration period, as well as the current period, in which research was resumed with less ambitious objectives. Thus, some considerations about the future of the Semantic Web Based Learning are presented.

The lack of human-friendly explanations from Artificial Intelligence (AI)—based decisions is a major concern for high stake applications. Towards mitigating the concern, Explainable AI (XAI) is an emerging area of research. In pre-modeling explainability, which is one of the notions of explainability, one can introduce explainability before training the model. One can also introduce explainability during training of the model or after training the model, in general, both of which are known as post-hoc explainability. Unfortunately, post-hoc explainability is not readily transparent and can be misleading, as it explains after the decision has been made and can be optimized to placate a subjective demand which is a form of bias. As a result, the explanation from it can be misleading even though it seems plausible. Explainability should incorporate knowledge from different domains such as philosophy, psychology, and cognitive science, so that the explanation is not just based on the researcher’s intuition of what constitutes a good explanation. Domain knowledge is an abstract, fuzzy, and high-level concept over the problem domain. For instance, in an image classification problem, the domain knowledge could be a dog has four legs, or a zebra has stripes, etc. However, the use of domain knowledge for explainability is under-focused and bound to problem-specific requirements. This chapter focuses on the notion of a pre-modeling explainability of AI-based “black box” models using domain knowledge. We demonstrate the collection and application of domain knowledge along with the quantification of explainability on an intrusion detection problem. Although AI-based Intrusion Detection Systems (IDS) provide accelerated speeds in intrusion detection, the response is still at a human speed when there is a human in the loop. The lack of explainability of an AI-based model is a key reason for this bottleneck as a human analyst has to understand the prediction before making the final decision towards mitigation of a problem. To mitigate this issue, in this chapter, we incorporate the CIA principle (i.e., domain knowledge) in an AI-based black box model for better explainability and generalizability of the model and demonstrate the process in detail. A major portion of this chapter is a compilation of our previously published work [1–4]. We start with a brief discussion of the problem, network intrusion detection and prevention, that we use to demonstrate our approach. Then we briefly introduced relevant domain knowledge and how it can be integrated into the overall architecture. In Sect. 3, we describe our experiments, followed by Sect. 4 which contains a discussion on results from the experiments. We conclude with limitations and future work in Sect. 5.

Within the educational context, researchers have focused on applying agent and ontology-based technologies to improve the processes of localization, retrieval, cataloging, and reuse of learning objects. This scenario highlights semantic heterogeneity issues, creating an excellent opportunity to evaluate, and explore ontology alignment techniques able to provide semantic integration between different ontologies. This work presents the MSSearch service, which combines state of the art agent and ontology-based technologies, with advanced alignment techniques to provide a semantic search service for a learning object repository. MSSearch was tested with a base of more than 11.000 learning object, answering queries in real-time. The quality of the answers were checked by educational experts and considered very satisfactory, when compared against similar queries made with the standard search engine of a public repository of learning objects, containing a similar set of learning objects.

The development of learning objects which are smaller chunks of learning content is often time consuming and expensive. Recent efforts in producing learning objects that are reusable and sharable were not very successful due to the lack of interoperability between different learning systems. Semantic Web technologies together with ontologies provide rich medium for facilitating e-learning via the semantic annotated learning objects and shared repositories. However, Semantic Web technologies have not yet been applied widely to deliver learning objects. This paper introduces enabling technologies for developing next-generation learning object repository for construction. Semantic Web, ontology, reasoners, and associated technologies are discussed within the paper. This paper also demonstrates how the ontologies and Semantic Web technologies can be used to develop and deliver intelligent, sharable and dynamic learning objects. The paper also outlines the development of repository using the ontology-driven Semantic Web approach and discusses the construction education ontology and semantic databases.

Deep learning based general language models have achieved state-of-the-art results in many popular tasks such as sentiment analysis and QA tasks. Text in domains like social media has its own salient characteristics. Domain knowledge should be helpful in domain relevant tasks. In this work, we devise a simple method to obtain domain knowledge and further propose a method to integrate domain knowledge with general knowledge based on deep language models to improve performance of emotion classification. Experiments on Twitter data show that even though a deep language model fine-tuned by a target domain data has attained comparable results to that of previous state-of-the-art models, this fine-tuned model can still benefit from our extracted domain knowledge to obtain more improvement. This highlights the importance of making use of domain knowledge in domain-specific applications.

E-Learning is a fast, just-in-time, learning process, which is becoming popular in distributed and dynamic environments such as the World Wide Web. The core of E-Learning is Learning Content Management System (LCMS). Traditional LCMS supports search of learning objects based on keywords. The conventional approach provides relevant as well as non-relevant results as they do not consider user's context. Also, precise search of non-textual resources such as audio-visual or multimedia based resources is not possible as these are not simple document files where keywords could be searched. The learner has to spend large amount of time in screening the learning resource to determine the best suiting his/her current learning requirements. In this paper an ontology based LCMS has been discussed. Ontology is a formal description of concepts in a real-world domain and helps to store the semantic information of that domain. It is more expressive than any other standard database model and facilitates knowledge sharing and reuse, i.e. a common understanding of various contents that reaches across people and applications. Developed LCMS incorporates semantic capabilities with the help of ontologies to enable semantic based contribution and retrieval of learning objects. Two very useful ontologies are created to describe the learner profile and the learning content. These ontologies have been used to classify the learning material, organize topics of a course and store the information pertaining to learner's profile. The proposed OLearner LCMS facilitates a resource contributor to contribute the resource to the OLearner portal with the metadata corresponding to the learning content and learner profile ontologies. These ontologies are further useful for a learner in searching a more precise and appropriate learning object.

During the last decade, increasing attention has been focused on ontologies and Ontological Engineering. The application of ontological engineering technique to e-learning, though still in its early stage, has become a phenomena tool in the integration and adaptation of a dynamic and flexible elearning environment, where tutors and learners are provided with the functionality of personalizing their learning process, this has not fully considered the generality of an entire field of study. This paper presents an ontology-based framework for classification of the entire courses of computer science at the university level. A model of an e-learning domain in the context of material classification that enables students retrieve information about a particular course in personalized ways was presented the method adopted elicited e-learning knowledge using documented materials, observation, consultation, prototyping among others. With domain expertise developed, the knowledge elicited was analyzed and formally represented using OWL-Description Logic. The design was implemented using Protégé 5.0 editor. The validation for accuracy and completeness was carried out with the domain and ontology experts using consistency checking reasoned, DL-Query. The results of the implementation show that elearning materials can be searched and retrieved easily and timely. The study also established and formalised an effective computational approach for representing information content of educational resources thereby helping to improve the flexibility of content representation in an e-learning system.Keywords: e-learning, Ontological Engineering, Ontology, Ontology Web Language (OWL), Knowledge Management, Description LogicVol. 26, No 1, June 2019

E-government faces huge challenges in achieving interoperability and integration, taking into account differences in laws, regulations, services, administrative processes and languages across regions and countries. On the other hand, issues like service, data and process integration have been researched by the Semantic Web community for several years now, and in the last two to three years we have witnessed the first applications of semantic technologies in real, operational e-government systems in both Europe and the US which address exactly these challenges. With this book, the editors present the latest research results on how to use semantic technologies in order to improve or even revolutionize the use of ICT in public administration systems. The contributions are organized into three parts: architectures and process integration, ontologies and interoperability, and portals and user interactions. They give a broad overview of how semantic technologies have been applied in different e-government projects funded from the European program for ICT Research and Development, and they cover a wide spectrum of semantic technologies such as development of domain and service ontologies, semantic enhancements of business process models, semantic Service-Oriented Architectures (SOAs) based on Semantic Web Services (SWS) frameworks, and ontology-based knowledge management. In this volume, researchers of Semantic Web technologies will find a wealth of challenging real-world scenarios to stimulate new fields of research, while developers of e-government systems as well as other stakeholders in public administration will appreciate the detailed presentations and discussions of numerous applications in areas such as e-government portals, personalization of Web-based public services, or integration and orchestration of public administration processes.

Structural Health Monitoring (SHM) is a condition-based maintenance technology using sensing systems. In SHM, the use of domain knowledge is essential: it motivates the use of machine learning approaches; it can be used to extract damage sensitive features and interpret the results by machine learning. This work focuses on two SHM problems: damage identification and substructure clustering. Our solutions to address them are based on machine learning techniques and robust feature extraction using domain knowledge. In the first problem, damage sensitive features were extracted using a frequency domain decomposition, followed by a robust one-class support vector machine for damage detection. In the second problem, a novel clustering technique and spectral moment feature were utilised for substructure grouping and anomaly detection. These methods were evaluated using data from lab-based structures and data collected from the Sydney Harbour Bridge. We obtained high damage detection accuracies and were able to assess damage severity. Furthermore, the clustering technique was able to group substructures of similar behaviour and detect spatial anomalies.

This paper discusses some important issues regarding the the management of Learning objects covering searching over repositories and different approaches of recommendation systems and presents a multiagent system based application model for indexing, retrieving and recommending learning objects stored in different and heterogeneous repositories. The objects within these repositories are described by filled fields using different metadata (data about data) standards. The searching mechanism covers several different learning object repositories and the same object can be described in these repositories by the use of different types of fields. Aiming to improve accuracy and coverage in terms of recovering a learning object and improve the relevance of the results we propose an information retrieval model based on a multiagent system approach and an ontological model to describe the covered knowledge domain.

Semantic technology is now taking shape in many forms and applications within the STM (scientific, technical and medical) industry. Understanding the difference between a Boolean search and semantic search is very important in determining what is best for your user community. We are going to explore the definition of semantic technology and its positive attributes and how it enhances scientific research. The aim is to provide the reader with a primer and definition of the semantic web technology, and it will provide an overview of Natural Language Processing (NLP) and the related tools (ontology, taxonomy and thesaurus) that are essential to creating a conceptual map of the data. With this foundation we will explore the semantic technology applications in the STM industry, in the commercial space, some of the exciting future possibilities for the Semantic Web and how a publisher integrates semantic technology into their applications and workflow tools for the author community.