Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Cited by Google
- Similars in SciELO
- Similars in Google
Share
DYNA
Print version ISSN 0012-7353On-line version ISSN 2346-2183
Abstract
PEREZ-MARTINEZ, Liz et al. Data mining algorithms for predicting the behavior of environmental indicators. Dyna rev.fac.nac.minas [online]. 2021, vol.88, n.219, pp.228-236. Epub Mar 15, 2022. ISSN 0012-7353. https://doi.org/10.15446/dyna.v88n219.95018.
The need to embrace adequate entrepreneurial focuses to achieve a better environmental performance constitutes an imminent task. Developing predictive models for environmental indicators constitutes the main objective of this paper. An information-technology tool that backs up the decision making will enable an efficient entrepreneurial environmental management, in such a way that they avoid errors that are commented at the present time. The application of techniques of data mining enabled capturing the last bosses and to reply to them, in addition to accomplish estimates with new data or out of sample, as well as inferring behaviors and future results, for the sake of anticipating possible situations of deterioration that compromise the environmental sustainability. The experiments designed to compare to the results in classification using the predictive models, prove that the percentage of error oscillates between the 4 % and the 5 %, what demonstrates extremely good degree of precision (height), according to the scales of verification.
Keywords : prediction; classification; ARIMA; temporal serie.