BUSCADOR RECOLECTA

Currently, the integration of technologies such as the Internet of Things and big data seeks to cover the needs of an increasingly demanding society that consumes more resources. The massification of these technologies fosters the transformation of cities into smart cities. Smart cities improve the comfort of people in areas such as security, mobility, energy consumption and so forth. However, this transformation requires a high investment in both socioeconomic and technical resources. To make the most of the resources, it is important to make prototypes capable of simulating urban environments and for the results to set the standard for implementation in real environments. The search for an environment that represents the socioeconomic organization of a city led us to consider universities as a perfect environment for small-scale testing. The proposal integrates these technologies in a traditional university campus, mainly through the acquisition of data through the Internet of Things, the centralization of data in proprietary infrastructure and the use of big data for the management and analysis of data. The mechanisms of distributed and multilevel analysis proposed here could be a powerful starting point to find a reliable and efficient solution for the implementation of an intelligent environment based on sustainability.

Currently, universities are being forced to change the paradigms of education, where knowledge is mainly based on the experience of the teacher. This change includes the development of quality education focused on students’ learning. These factors have forced universities to look for a solution that allows them to extract data from different information systems and convert them into the knowledge necessary to make decisions that improve learning outcomes. The information systems administered by the universities store a large volume of data on the socioeconomic and academic variables of the students. In the university field, these data are generally not used to generate knowledge about their students, unlike in the business field, where the data are intensively analyzed in business intelligence to gain a competitive advantage. These success stories in the business field can be replicated by universities through an analysis of educational data. This document presents a method that combines models and techniques of data mining within an architecture of business intelligence to make decisions about variables that can influence the development of learning. In order to test the proposed method, a case study is presented, in which students are identified and classified according to the data they generate in the different information systems of a university.

The large volumes of data that exist in the universities keep important information of each student. Analyzing this data represents a challenge for data scientists due to the number of resources they consume. Many of the universities do not have the capacity of infrastructure as well as human resources to do it for this reason they desist from the analysis of data depriving themselves of generating knowledge about their students. The range of sensors that generate data in a university is so wide that doing an analysis of data through a traditional method such as business intelligence does not provide accurate results and their response times are not as expected. This work proposes the use of big data techniques in a university to obtain accurate results in real time that will help in making decisions improving education and learning.

Currently, the 2019 Coronavirus Disease pandemic has caused serious damage to health throughout the world. Its contagious capacity has forced the governments of the world to decree isolation and quarantine to try to control the pandemic. The consequences that it leaves in all sectors of society have been disastrous. However, technological advances have allowed people to continue their different activities to some extent while maintaining isolation. Universities have great penetration in the use of technology, but they have also been severely affected. To give continuity to education, universities have been forced to move to an educational model based on synchronous encounters, but they have maintained the methodology of a face-to-face educational model, what has caused several problems in the learning of students. This work proposes the transition to a hybrid educational model, provided that this transition is supported by data analysis to identify the new needs of students. The knowledge obtained is contrasted with the performance presented by the students in the face-to-face modality and the necessary parameters for the transition to this modality are clearly established. In addition, the guidelines and methodology of online education are considered in order to take advantage of the best of both modalities and guarantee learning.