Tool for the extraction and analysis of information obtained from the social network Twitter, as support for the procedures: new qualified registry and renewal of records

Abstract

This article is derived from the research project that is being developed within the systems engineering program of the National Open and Distance University UNAD, which describes the preliminary advances in the development of the software tool called ARS-SIAVA, which allow the extraction of data from social networks, to identify in a timely manner the technological trends in academic training, behaviors of the labor market in the areas of IT information technologies both in regional, national and international contexts; by collecting and analyzing information obtained from the Twitter social network. Allowing this input to be a basis for the self-assessment processes in the renewal and obtaining of new qualified records by the Ministry of National Education, especially in support of the justification of the programs that correspond to condition 2 of the established quality criteria by the Ministry of Education in obtaining qualified records. In the development of the research two aspects were considered: A first part where text mining is carried out, which obtains data through an API, data that is collected in the form of tweets, imported into CSV format which is understandable for the R programming language, which previously goes through a debugging and adaptation process according to the requirements of this language. The second part corresponds to the analysis of feelings, for which there is an algorithm to which a manual training is carried out that through Machine Learning learns and finally is capable of predicting trends, which are projected in graphics, word clouds as well of frequency tables and visualization of statistical data that can contribute to decision-making.