Implementación de un Modelo de Regresión Lineal Múltiple aplicando (PLN) para predicción de artículos científicos.

Abstract

This article explains the importance of multiple linear regression in different applications, such as natural language processing and time series forecasting. Its relevance in the creation of intelligent tutoring systems that adapt to the individual needs of students is highlighted. The theoretical framework on regression models is presented, from simple linear regression to multiple linear regression, and the different types of variables involved in the model are described. In addition, the types of data, data preparation techniques, model evaluation, and methodology used in data analysis work are discussed. The crisp-dm methodology was applied, which is divided into six phases: collect data, prepare the data, model, evaluate, implement and maintain. The process of data collection and labeling of the IoT data, the loading and visualization of the database, and the techniques used in the cleaning and transformation of variables are described. Some common data preparation techniques and measures used to assess the quality of model fit are also explained.