Bernardo Pereira Nunes

Interlinking educational resources and the web of data: A survey of challenges and approaches

Purpose - Research in the area of 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, or interface mechanisms. More recently, semantic technologies were taken into account to improve interoperability. The linked data approach has emerged as the de facto standard for sharing data on the web. To this end, it is obvious that the application of linked data principles offers a large potential to solve interoperability issues in the field of TEL. This paper aims to address this issue. Design/methodology/approach - In this paper, approaches are surveyed that are aimed towards a vision of linked education, i.e. education which exploits educational web data. It particularly considers the exploitation of the wealth of already existing TEL data on the web by allowing its exposure as linked data and by taking into account automated enrichment and interlinking techniques to provide rich and well-interlinked data for the educational domain. Findings - So far web-scale integration of educational resources is not facilitated, mainly due to the lack of take-up of shared principles, datasets and schemas. However, linked data principles increasingly are recognized by the TEL community. The paper provides a structured assessment and classification of existing challenges and approaches, serving as potential guideline for researchers and practitioners in the field. Originality/value - Being one of the first comprehensive surveys on the topic of linked data for education, the paper has the potential to become a widely recognized reference publication in the area.

Combining a co-occurrence-based and a semantic measure for entity linking

One key feature of the Semantic Web lies in the ability to link related Web resources. However, while relations within particular datasets are often well-defined, links between disparate datasets and corpora of Web resources are rare. The increasingly widespread use of cross-domain reference datasets, such as Freebase and DBpedia for annotating and enriching datasets as well as documents, opens up opportunities to exploit their inherent semantic relationships to align disparate Web resources. In this paper, we present a combined approach to uncover relationships between disparate entities which exploits (a) graph analysis of reference datasets together with (b) entity co-occurrence on the Web with the help of search engines. In (a), we introduce a novel approach adopted and applied from social network theory to measure the connectivity between given entities in reference datasets. The connectivity measures are used to identify connected Web resources. Finally, we present a thorough evaluation of our approach using a publicly available dataset and introduce a comparison with established measures in the field.

A Scalable Approach for Efficiently Generating Structured Dataset Topic Profiles

The increasing adoption of Linked Data principles has led to an abundance of datasets on the Web. However, take-up and reuse is hindered by the lack of descriptive information about the nature of the data, such as their topic coverage, dynamics or evolution. To address this issue, we propose an approach for creating linked dataset profiles. A profile consists of structured dataset metadata describing topics and their relevance. Profiles are generated through the configuration of techniques for resource sampling from datasets, topic extraction from reference datasets and their ranking based on graphical models. To enable a good trade-off between scalability and accuracy of generated profiles, appropriate parameters are determined experimentally. Our evaluation considers topic profiles for all accessible datasets from the Linked Open Data cloud. The results show that our approach generates accurate profiles even with comparably small sample sizes (10%) and outperforms established topic modelling approaches.

Identifying candidate datasets for data interlinking

One of the design principles that can stimulate the growth and increase the usefulness of the Web of data is URIs linkage. However, the related URIs are typically in different datasets managed by different publishers. Hence, the designer of a new dataset must be aware of the existing datasets and inspect their content to define sameAs links. This paper proposes a technique based on probabilistic classifiers that, given a datasets S to be published and a set T of known published datasets, ranks each Ti ∈ T according to the probability that links between S and Ti can be found by inspecting the most relevant datasets. Results from our technique show that the search space can be reduced up to 85%, thereby greatly decreasing the computational effort.

Combining a Co-occurrence-Based and a Semantic Measure for Entity Linking

One key feature of the Semantic Web lies in the ability to link related Web resources. However, while relations within particular datasets are often well-defined, links between disparate datasets and corpora of Web resources are rare. The increasingly widespread use of cross-domain reference datasets, such as Freebase and DBpedia for annotating and enriching datasets as well as documents, opens up opportunities to exploit their inherent semantic relationships to align disparate Web resources. In this paper, we present a combined approach to uncover relationships between disparate entities which exploits (a) graph analysis of reference datasets together with (b) entity co-occurrence on the Web with the help of search engines. In (a), we introduce a novel approach adopted and applied from social network theory to measure the connectivity between given entities in reference datasets. The connectivity measures are used to identify connected Web resources. Finally, we present a thorough evaluation of our approach using a publicly available dataset and introduce a comparison with established measures in the field.

Recommending Tripleset Interlinking through a Social Network Approach

Tripleset interlinking is one of the main principles of Linked Data. However, the discovery of existing triplesets relevant to be linked with a new tripleset is a non-trivial task in the publishing process. Without prior knowledge about the entire Web of Data, a data publisher must perform an exploratory search, which demands substantial effort and may become impracticable, with the growth and dissemination of Linked Data. Aiming at alleviating this problem, this paper proposes a recommendation approach for this scenario, using a Social Network perspective. The experimental results show that the proposed approach obtains high levels of recall and reduces in up to 90% the number of triplesets to be further inspected for establishing appropriate links.

Two Approaches to the Dataset Interlinking Recommendation Problem

Whenever a dataset t is published on the Web of Data, an exploratory search over existing datasets must be performed to identify those datasets that are potential candidates to be interlinked with t. This paper introduces and compares two approaches to address the dataset interlinking recommendation problem, respectively based on Bayesian classifiers and on Social Network Analysis techniques. Both approaches define rank score functions that explore the vocabularies, classes and properties that the datasets use, in addition to the known dataset links. After extensive experiments using real-world datasets, the results show that the rank score functions achieve a mean average precision of around 60%. Intuitively, this means that the exploratory search for datasets to be interlinked with t might be limited to just the top-ranked datasets, reducing the cost of the dataset interlinking process.

A Topic Extraction Process for Online Forums

Forums play a key role in the process of knowledge creation, providing means for users to exchange ideas and to collaborate. However, educational forums, along several others online educational environments, often suffer from topic disruption. Since the contents are mainly produced by participants (in our case learners), one or a few individuals might change the course of the discussions. Thus, realigning the discussed topics of a forum thread is a task often conducted by a tutor or moderator. In order to support learners and tutors to harmonically align forum discussions that are pertinent to a given lecture or course, in this paper, we present a method that combines semantic technologies and a statistical method to find and expose relevant topics to be discussed in online discussion forums.

Educational Forums at a Glance: Topic Extraction and Selection

Web forums play a key role in the process of knowledge creation, providing means for users to exchange ideas and to collaborate. However, educational forums, along several others online educational environments, often suffer from topic disruption. Since the contents are mainly produced by participants (in our case learners), one or few individuals might change the course of the discussions. Thus, realigning the discussed topics of a forum thread is a task often conducted by a tutor or moderator. In order to support learners and tutors to harmonically align forum discussions that are pertinent to a given lecture or course, in this paper, we present a method that combines semantic technologies and a statistical method to find and expose relevant topics to be discussed in online discussion forums. We surveyed the outcomes of our topic extraction and selection method with students, professors and university staff members. Results suggest the potential usability of the method and the potential applicability in real learning scenarios.

TRTML - A Tripleset Recommendation Tool Based on Supervised Learning Algorithms

The Linked Data initiative promotes the publication of interlinked RDF triplesets, thereby creating a global scale data space. However, to enable the creation of such data space, the publisher of a tripleset \(t\) must be aware of other triplesets that he can interlink with \(t\). Towards this end, this paper describes a Web-based application, called TRTML, that explores metadata available in Linked Data catalogs to provide data publishers with recommendations of related triplesets. TRTML combines supervised learning algorithms and link prediction measures to provide recommendations. The evaluation of the tool adopted as ground truth a set of links obtained from metadata stored in the DataHub catalog. The high precision and recall results demonstrate the usefulness of TRTML.