Could not download file: This paper is available to authorised users only.


R. Llorente, M. Morant

Universidad Politecnica de Valencia (SPAIN)
This paper proposes and demonstrates, for the first time to our knowledge, the application of deep data-mining techniques in practical lessons applied to the laboratory practice work done in the subject “Análisis de Sistemas Contínuos” in the second term of the first year of the Telecommunication Engineering Degree studies at the Escuela Politécnica Superior de Gandia from the Universidad Politecnica de Valencia, Spain.
Tailored deep data-mining [1] is applied here to track the knowledge construction process of the alumni. Data-mining is applied over a large set of variables which are tracked during a simple on-line exercise done at the end of the laboratory lesson. The main variables considered for deep data-mining that are summarised in:

Ts: Time spent by the student observing the question on-screen
Tr: Times the studes has recalled the question to be presented in the screen
Ai: Number of times the studies has changed the answer
Seq_a: Sequence of answers selected by the student when changing the answer of a given question
Seq:q: Sequence of questions requested by the student after the first presentation
S1… SN: Individual scores per question
Lq: Last question answered by the student
S: Final score

The correlation of the different variables is analysed in order to identify the bottlenecks or roadblocks in the student learning process. This is taken into account in the web application for laboratory evaluation purposes dealing with Fourier and Laplace domains in Telecommunication engineering. This on-line application includes a “control panel” of the subject which is presented in real-time to the lecturer during the realisation of the on-line exercise. This “control panel” (CP) is a translation of the business control panel that can be found in advanced business administration techniques [2].
The special importance of the proposed technique, is that data-mining is done at class-level and also at student-level, which permits the proper guide of the knowledge building process at personal level. This fully tailored tracking is in line with the personalised education guidelines present in all major high-level education institutions.
The results shown in this paper reflect the fist experience of the implementation of a CP in practical work lesson in Engineering studies. The implementation work is based on an Oracle database with ad-hoc programming. After the development of the data-mining tool and the introduction of this panel, and the associate corrective lecturing actions, the pass-rate of the laboratory work in the subject has increased by in 37% the first year it was introduced.

[1] Ian H. Witten, Eibe Frank, “Data Mining: Practical Machine Learning Tools and Techniques”, Ed. Morgan Kaufman, 2005.
[2] Adriana Di Liberto, Roberto Mura, Francesco Pigliaru “How to Measure the Unobservable: A Panel Technique for the Analysis of TFP Convergence”, FEEM Working Paper No. 16.05, January 2005.