Intervention Model with Data Mining Techniques to Work with Dating Violence Victims Using a Social Support Network
摘要
Dating violence is a problem that affects a large proportion of young women in Mexico, so it is necessary to design and implement policies to support its victims. Although progress has been made in this area, technological and data mining solutions can help to eradicate the problem. Therefore, the purpose of this chapter is to propose an automated violence assessment process, through the design and implementation of a technological intervention model in RedSIAM, in order to provide users with appropriate care according to the type and level of violence detected. Subsequently, the risk factors for violence are evaluated and data mining techniques are used to propose the model. Through processing with clustering algorithms, violence profiles are created according to the characteristics of the users. Later, the resulting information is used to document, generate behavioral patterns and classify each user into one of the profiles already identified. This processing is done using classification algorithms such as EM (Expectation-Maximization) and Simple K-Means. Within the social network, each profile has access to different content depending on the group it belongs to. The intervention model allows us to have analyzable data to understand the patterns of violence among the victims, in order to help the social workers in their decision-making task to provide them with timely assistance.