Aplicación de minería de datos para determinar las causas de la violencia de género contra las mujeres en Ecuador.
Gender violence against women is a problem that is increasing every day worldwide, this titling work focused on cases registered in Ecuador, therefore, was selected the set of data collected by INEC through the National Survey of Family Relations and Gender Violence against women, which rests on its...
Saved in:
| Main Author: | |
|---|---|
| Format: | bachelorThesis |
| Language: | spa |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://dspace.unl.edu.ec/jspui/handle/123456789/22773 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Add Tag | Summary: | Gender violence against women is a problem that is increasing every day worldwide, this titling work focused on cases registered in Ecuador, therefore, was selected the set of data collected by INEC through the National Survey of Family Relations and Gender Violence against women, which rests on its website. Data mining techniques were applied to this database to determine the main causes for which a woman suffers violence, including intimate partner violence, dating violence, violence in childhood as opposed to intimate partner violence, and violence in the social, labor and educational spheres. Taking into account each of these areas, the database was fragmented, obtaining thirty subsets of data-oriented to each type of violence, the area in which it occurred and considering the marital status of women. Four types of violence were defined: psychological, physical, economic or patrimonial and sexual violence. Six classification algorithms were also selected for data mining: J48, exhaustive CHAID, neural networks, IBk, decision table, and Random forest; the software tools used were RStudio, SPSS Statistics, and Weka. In some cases, it was necessary to apply the SMOTE algorithm to the data sets to optimize the results. The methodology used consists of five phases: information search, data collection, database cleaning, application of data mining techniques and interpretation of results. The results obtained were presented according to the established objectives, starting with the identification of the repositories where information on reported cases of violence was found stored, then data mining techniques were applied to the subsets of data, culminating with the analysis and presentation of the results obtained. Finally, in the discussion section, the development of an alternative proposal and the technical-economic-environmental assessment were presented. |
|---|