论文信息 - Data Wrangling: A Decisive Step for Compact Regression Trees

Data Wrangling: A Decisive Step for Compact Regression Trees

Nowadays, modern visualization and decision support platforms provide advanced and interactive tools for data wrangling, in order to facilitate data analysis. Nonetheless, it is a tedious process that requires a deep experience in data transformation. In this paper, we propose an automated data wrangling method, based on a genetic algorithm, that helps to obtain simpler regression trees.

Thomas Tamisier | Olivier Parisot | Yoanne Didry

[1] Robert Engels,et al. Using a Data Metric for Preprocessing Advice for Data Mining Applications , 1998, ECAI.

[2] Leo Breiman,et al. Classification and Regression Trees , 1984 .

[3] Jeffrey Heer,et al. Wrangler: interactive visual specification of data transformation scripts , 2011, CHI.

[4] Shouhong Wang,et al. Mining Data Quality In Completeness , 2007, ICIQ.

[5] Thomas Tamisier,et al. User-Driven Data Preprocessing for Decision Support , 2013, CDVE.

[6] Dominique Brodbeck,et al. Research directions in data wrangling: Visualizations and transformations for usable and credible data , 2011, Inf. Vis..

[7] Thomas Bäck,et al. A Survey of Evolution Strategies , 1991, ICGA.

[8] Ian H. Witten,et al. Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[9] Sotiris B. Kotsiantis,et al. Decision trees: a recent overview , 2011, Artificial Intelligence Review.