Tuesday, January 11, 2011

The gist of the first reading

The following are the gist my eyes could catch while skimming through the journal by Baker, S.J.D., Yacef, K. (2009) The State of Educational Data Mining in 2009: A Review and Future Visions: http://www.educationaldatamining.org/JEDM/images/articles/vol1/issue1/JEDMVol1Issue1_BakerYacef.pdf

My short review is laid out at the end...

The Educational Data Mining community website, http://www.educationaldatamining.org/ , defines educational data mining as follows: “Educational Data Mining is an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and using those methods to better understand students, and the settings which they learn in.

Another name for data mining is Knowledge Discovery in Databases (KDD) - it is the field of discovering novel and potentially useful information from large amounts of data [Witten and Frank 1999].

Baker [in press] classifies work in educational data mining as follows:

  1. Prediction
    - Classification
    - Regression
    - Density estimation
  2. Clustering
  3. Relationship mining
    - Association rule mining
    - Correlation mining
    - Sequential pattern mining
    - Causal data mining
  4. Distillation of data for human judgment
  5. Discovery with models

Student models represent information about a student’s characteristics or state, e.g. the student’s current knowledge, motivation, meta-cognition, and attitudes. Modeling student individual differences in these areas enables software to respond to those individual differences, significantly improving student learning [Corbett 2001].

As Bartneck and Hu [2009] have noted, Google Scholar is the most comprehensive source for citations – particularly for the conferences which are essential for understanding Computer Science research.

Recent years have also seen major changes in the types of EDM methods that are used, with prediction and discovery with models increasing while relationship mining becomes rarer. How would these trends shift in the years to come?

Educational data mining methods have had some level of impact on education and related interdisciplinary fields (e.g. artificial intelligence in education, intelligent tutoring systems, and user modeling).

Basically this journal talks about the literatures written in the topic of Educational Data Mining (EDM), and how the trend in research has shifted from one aspect to another. Initially, the method used for EDM was more on relationship mining, but towards the later stage the researches are using more of prediction and discovery with models as methods.

One part that interests me in this article is the impact shown on EDM on related fields such as Artificial Intelligence (AI), which is what I'm into in this past 1 year. In some ways, I have a 'hunch' and prediction that I may be doing some research on this area with the connection to AI.

Start the ball rolling,
- Shazz @ LAK
11 Jan 2011

No comments:

Post a Comment