Improving students’ satisfaction in a massive online academic writing course

Proceedings of the International Conference on Advanced Learning Technologies ICALT 2011, Athens, USA

Teachers in fully online courses face the challenge of achieving students’ satisfaction – usually measured using students’ surveys – as it is known to affect engagement and motivation, which in turn improves learning outcomes. This challenge grows in complexity with the number of students enrolled in the course. In this paper we report an attempt to improve students’ perception on their teacher in a massive online academic writing course. In the experimental group the way contents were delivered was changed, the same as the role teachers had. Content was embedded in a story and the teacher acted as the presenter of each chapter during the course. Results showed that students’ perception improved significantly, particularly on how they perceived the teacher’s availability.

Concept maps as cognitive visualisations of writing assignments

International Journal of Educational Technology & Society, 2011

Writing assignments are ubiquitous in higher education. Writing develops not only communication skills, but also higher-level cognitive processes that facilitate deep learning. Cognitive visualizations, such as concept maps, can also be used as part of learning activities including as a form of scaffolding, or to trigger reflection by making conceptual understanding visible at different stages of the learning process. We present Concept Map Miner (CMM), a tool that automatically generates Concept Maps from students’ compositions, and discuss its design and implementation, its integration to a writing support environment and its evaluation on a manually annotated corpora of university essays (N=43). Results show that complete CM, with concepts and labeled relationships, are possible and its precision depends the level of summarization (number of concepts) chosen.

Analysis of a Gold Standar for Concept Map Mining – How humans summarize text using Concept Maps

Essay writing and concept mapping are both learning activities that involve higher level thinking, moreover the latter has been used to support writing by presenting a different visualization of the essay, facilitating student’s reflection. However as concept mapping is a time consuming task, the immediacy of such feedback is impossible. Concept Map Mining (CMM) is the automatic extraction of concept maps from essays, which would allow immediate feedback on writing activities. CMM is currently work in progress in an advanced stage, and its evaluation requires a gold standard of concept maps extracted from essays by human annotators. This paper reports on the creation of such a gold standard, and analyzes patterns that will help understanding how humans summarize text using concept maps. Such patterns will inform the design of the CMM algorithms. This analysis shows that several interesting patterns arise when humans face the task of extracting concept maps from essays.

Glosser: Enhanced Feedback for Student Writing Tasks

Proceedings of the International Conference on Advanced Learning Technologies ICALT 2008

We describe Glosser, a system that supports students in writing essays by 1) scaffolding their reflection with trigger questions, and 2) using text mining techniques to provide content clues that can help answer those questions.
A comparison with other computer generated feedback and scorings systems is provided to explain the novelty of the approach. We evaluate the system with Wiki pages produced by postgraduate students as part of their assessment.

Concept Map Mining: A definition and a framework for its evaluation

Proceedings of the IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology WI-IAT 2008

Concept maps are visual representations of knowledge,
widely used in educational contexts. We use the term ”Concept Map Mining” (CMM) to refer to the automatic extraction of Concept Maps from documents such as essays. The principles behind CMM have been proposed for applications such as: information extraction in specific knowledge domains, the measurement of student understanding and misconceptions based on written essays, and as a preliminary step to creating domain ontologies.
Previous work on the automatic extraction of concept maps present two problems: 1) overly simplistic and varying definitions of concept maps, and 2) the lack of an evaluation framework that can be used to measure the quality of the generated maps. In this paper, we propose a formal definition of the term CMM, with a focus on educational applications. We also propose an evaluation framework that will allow other researchers to share a common ground to evaluate the performance of CMM methods.

Concept Extraction from student essays, towards Concept Map Mining

This paper presents a new approach for automatic concept extraction, using grammatical parsers and Latent Semantic Analysis. The methodology is described, also the tool used to build the benchmarking corpus. The results obtained on student essays shows good inter-rater agreement and promising machine extraction performance. Concept extraction is the first step to automatically extract concept maps from student’s essays or Concept Map Mining.

Single document semantic spaces

to appear in Proceedings of the Australian Joint Conference on Artificial Intelligence 2009

Latent Semantic Analysis (LSA) has been successfully used in a number of information retrieval, document visualization and summarization applications. LSA semantic spaces are normally created from large corpora that reflect an assumed background knowledge. However the right size and coverage of the background knowledge for each application are still open research questions. Moreover, LSA's computational cost is directly related to the size of the corpus, making the technique inviable in many cases.
This paper introduces a technique for creating semantic spaces using a single document and no background knowledge, which cuts computational cost and is domain independent. Single document semantic spaces' reliability was evaluated on a collection of student essays. Several semantic spaces generated from large corpora and single documents were used to compare how essays are represented. The distance between consecutive sentences in the essays changes between semantic spaces, but the rank of the distances is preserved. The results show that high correlations (0.7) of ranked distances between sentences can be achieved on the different spaces for the weight schemes evaluated.
This has important implications for the applications discussed.