LILE 2019 is collocated with ACM WebSci 2019. Building on the previous seven editions and its growing community, LILE2019 will provide an interdisciplinary forum for researchers and practitioners who make innovative use of Web data for educational purposes, spanning areas such as learning analytics, Web mining, data and Web science, psychology and the social sciences. The previous editions of the LILE workshop were successfully held at the ESWC, WWW, ISWC and WebSci conferences. LILE2019 targets a public full day workshop that consists of keynotes, presentations of accepted papers and posters and discussion. The intended audience consists of researchers and practitioners from the general areas of web science, learning analytics, psychology, computational social science or the social sciences.
Distance teaching and openly available educational resources on the Web are becoming common practices with public higher education institutions as well as private training organizations. In addition, informal learning and knowledge exchange are inherent to the daily online interactions, when searching the Web or using learning and knowledge-related social networks, such as Bibsonomy, Slideshare, Wikipedia or Videolectures, or general purpose social environments, such as LinkedIn, where matters related to skills, competence development or training are central concerns of involved stakeholders. These interactions generate a vast amount of data, about informal knowledge resources of varying granularity as well as user activities, including informal indicators for learning and competences.
On the other hand, the prevalence of entity-centric Web data, facilitated through Open Data, Knowledge Graphs or Linked Data, as well as the more recent widespread adoption of embedded annotations through schema.org, Microdata and RDFa has led to the availability of vast amounts of semi-structured data which facilitates interpretation and reuse of Web content and data in learning scenarios.
Initiatives such as LinkedUp or the more recent AFEL project2 have already made available collections of learning-related data, covering both user activity as well as resource-centric information. The widespread analysis of both informal and formal learning activities and resources has the potential to fundamentally aid and transform the production, recommendation and consumption of learning services and content. Typical scenarios include the use of machine learning to automatically classify learning performance, competences or user knowledge by learning from the vast amounts of available data or to exploit resource-centric data and knowledge graphs to automatically generate learning resources or assessment items.
However, interpreting learning activities and online interactions requires a highly interdisciplinary skillset, involving knowledge about learning theory, psychology and sociology as well as technical means to enable data analysis in large-scale heterogeneous data. Building on the success of several previous editions, LILE2019 aims at addressing such challenges by providing a forum for researchers and practitioners who make innovative use of Web data for educational purposes, spanning areas such as learning analytics, Web mining, data and Web science, psychology and the social sciences.