The present study is a trial in developing an automatic computerized coding framework with text mining techniques to identify the characteristics of redemption and contamination in life narratives written by undergraduate students. In the initial stage of text classification, the keyword-based classifier algorithm – product score model – performed more sensitive in finding the “change” elements in life stories comparing to the Decision Tree and Naïve Bayes models. The verbal features of life narratives were also discussed to enhance the classification accuracy further in assessing students’ personality adaption.
|Title of host publication||EDM 2011: 4th International Conference on Educational Data Mining|
|Subtitle of host publication||Proceedings, Eindhoven, July 6-8, 2011|
|Editors||Mykola Pechenizkiy, Toon Calders, Cristina Conati, Sebastian Ventura, Cristobal Romero, John Stamper|
|Place of Publication||Eindhoven|
|Publication status||Published - 6 Jul 2011|
|Event||4th International Conference on Educational Data Mining, EDM 2011 - Eindhoven, Netherlands|
Duration: 6 Jul 2011 → 8 Jul 2011
Conference number: 4
|Conference||4th International Conference on Educational Data Mining, EDM 2011|
|Period||6/07/11 → 8/07/11|
He, Q., Veldkamp, B. P., & Westerhof, G. J. (2011). Computerized coding system for life narratives to assess students' personality adaption. In M. Pechenizkiy, T. Calders, C. Conati, S. Ventura, C. Romero, & J. Stamper (Eds.), EDM 2011: 4th International Conference on Educational Data Mining: Proceedings, Eindhoven, July 6-8, 2011 (pp. 325-326). Eindhoven.