Competence is a human potential for action, leading to performance, when put to action. With tracking, processing, and analysis technology, traces of performance allow prediction of and reasoning about underlying competence.
We deploy data science techniques and technologies to systematically develop a rich toolkit for human performance extraction, analysis, and prediction. In a sense, in this area we do data science with a specialisation in knowledge media.
The WAPLA workshop aims at analyzing with an applied and practical perspective different issues and challenges related to learning analytics. Learning analytics can be broadly defined as the methods and techniques to reach conclusions about the learning process in order to improve it. First of all, there is a need of collecting data since you cannot improve what you do not monitor. Nowadays, there are proper technical tools to collect and retrieve all the interactions of the users with the different learning resources and activities. The retrieve of these data enables its analysis and the application of learning analytics for improving the learning process. Scenarios such as MOOCs (Massive Open Online Courses) provides a good case study as it makes even more necessary the application of learning analytics techniques because there is a need of tools for teachers to easily monitor the learning process and students for receiving automatic feedback and awareness. The correct practical application of learning analytics can bring a lot of advantages. Many issues should be considered for a successful application of learning analytics in courses and in the organizations as a whole.
Click here for more.
“Learning Analytics in R with SNA, LSA, and MPIA” by Fridolin Wild published by Springer. Summary: This book introduces Meaningful Purposive Interaction Analysis (MPIA) theory, which combines social network analysis (SNA) with latent semantic analysis (LSA) to help create and analyse a meaningful learning landscape from the digital traces left by a learning community in the co-construction of knowledge.
The hybrid algorithm is implemented in the statistical programming language and environment R, introducing packages which capture – through matrix algebra – elements of learners’ work with more knowledgeable others and resourceful content artefacts. The book provides comprehensive package-by-package application examples, and code samples that guide the reader through the MPIA model to show how the MPIA landscape can be constructed and the learner’s journey mapped and analysed. This building block application will allow the reader to progress to using and building analytics to guide students and support decision-making in learning.
Dr. Fridolin Wild (Senior Research Fellow and Scientific Director of WEKIT) spoke about Wearables Enhanced Knowledge Intensive Training (WEKIT). Augmented reality and wearables are potential game changers, re-inventing the way we perceive and work with information. In the publicly-funded WEKIT project (Horizon 2020), we explore experience capturing with AR and wearables to support on the one hand observation of a master performing problem-solving tasks and, on the other, to deliver augmented real-time guidance to trainees. In this Research Centre talk, Fridolin introduced the department to the project and outlined the future R&D timeline, and paved the way for further collaboration on AR and wearables within the department.
ReflectR is a tool for researchers who are interested in analysing text on its reflective capabilities. Traditionally, such coding is done using human experts – a work intense and hence costly process. ReflectR can help do this more efficiently, with almost the same accuracy (see upcoming publication of thesis of Thomas Ullmann): it can compare your input with its internal collection of reflective texts and give you instant feedback on whether your sentences are reflective or rather descriptive.
Publication: Thomas Daniel Ullmann (2015): Automated detection of reflection in texts. A machine learning based approach, PhD thesis, The Open University, link.
’Learning’ means building competence through (targeted) conversation between an aspirant and a typically more experienced other. Such conversation, – particularly when born digital in form of messages, essays, articles, or books available online –, can be subjected to analysis, looking into both the developed conceptual structures as well as the social interaction leading to them.
Meaningful Purposive Interaction Analysis (MPIA) was developed to facilitate the analysis of social appropriation of meaning. It uses conceptual spaces constructed from the conversational content collected from networked learning activity in order to represent persons and their competence demonstrated.
MPIA is dervied from the same mathematical foundation as latent semantic analysis and (social) network analysis, sharing their advantages, while at the same time overcoming some of their short- comings.
The package can be downloaded here:
File: mpia_0.71.tar.gz (14.4 MB)
The basic idea of latent semantic analysis (lsa) is, that text do have a higher order (=latent semantic) structure which, however, is obscured by word usage (e.g. through the use of synonyms or polysemy). By using conceptual indices that are derived statistically via a truncated singular value decomposition (a two-mode factor analysis) over a given document-term matrix, this variability problem can be overcome.
Publication: Fridolin Wild, Christina Stahl (2007): Investigating unstructured texts with latent semantic analysis, In: Decker, Lenz (Eds.): Studies in Classification, Data Analysis, and Knowledge Organization, Springer, pp. 383-390, link.
PAL maintains the R task view on natural language processing, watching over the packages in the scope and helping with further development.
Natural language processing has come a long way since its foundations were laid in the 1940s and 50s (for an introduction see, e.g., Jurafsky and Martin (2008): Speech and Language Processing, Pearson Prentice Hall). The CRAN task view collects relevant R packages that support computational linguists in conducting analysis of speech and language on a variety of levels – setting focus on words, syntax, semantics, and pragmatics.
In recent years, within this R community, we have elaborated a framework to be used in packages dealing with the processing of written material: the package tm. Extension packages in this area are highly recommended to interface with tm’s foundational routines and useRs are encouraged to join in the discussion on further developments of this framework package. To get into natural language processing, the cRunch service and its collection of tutorials may be helpful.
Champion: Fridolin Wild
cRunch is a service and an infrastructure for computationally-intense learning analytics. It supports researchers in munching away the data points, generated by learners online in the co-construction of knowledge – helping learners, faculty, and administration with living reports, created live from data.
cRunch currently has three components: the studio, the reports plus services, and the sharing facilities. The studio is a browser-based work space for exploratory data programming, providing the famous Rstudio server as a hosted service. The studio uses the simple to learn statistical programming language R – for data manipulation – and a wiki-like markup language called ‘markdown’ – for reports. The reports, however, are more than just text. We even call them ‘living documents’, as in addition to static text they also access live data executing live analytical scripts. cRunch services are kind of more complex reports, some of them intended for other machines rather than humans. Finally, the sharing facilities are the interface to the world. They help the cRunch community to remix, recycle, repurpose data and code.