Vera Miguéis

Abstract
The present study puts forward a regression analytic model based on the random forest algorithm, developed to predict, at an early stage, the global academic performance of the undergraduates of a polytechnic higher education institution. The study targets the universe of an institution composed of 5 schools rather than following the usual procedure of delimiting the prediction to one single specific degree course. Hence, we intend to provide the institution with one single tool capable of including the heterogeneity of the universe of students as well as educational dynamics. A different approach to feature selection is proposed, which enables to completely exclude categories of predictive variables, making the model useful for scenarios in which not all categories of data considered are collected. The introduced model can be used at a central level by the decision-makers who are entitled to design actions to mitigate academic failure. © 2019, Springer Nature Switzerland AG.

Abstract
In the increasingly competitive fashion retail industry, companies are constantly adopting strategies focused on adjusting the products characteristics to closely satisfy customers' requirements and preferences. Although the lifecycles of fashion products are very short, the definition of inventory and purchasing strategies can be supported by the large amounts of historical data which are collected and stored in companies' databases. This study explores the use of a deep learning approach to forecast sales in fashion industry, predicting the sales of new individual products in future seasons. This study aims to support a fashion retail company in its purchasing operations and consequently the dataset under analysis is a real dataset provided by this company. The models were developed considering a wide and diverse set of variables, namely products' physical characteristics and the opinion of domain experts. Furthermore, this study compares the sales predictions obtained with the deep learning approach with those obtained with a set of shallow techniques, i.e. Decision Trees, Random Forest, Support Vector Regression, Artificial Neural Networks and Linear Regression. The model employing deep learning was found to have good performance to predict sales in fashion retail market, however for part of the evaluation metrics considered, it does not perform significantly better than some of the shallow techniques, namely Random Forest.

Abstract
This paper presents a methodology based on random forest algorithm to predict the undergraduate academic performance of students from a polytechnic institution. The approach followed enabled to select 11 explanatory variables, starting from an initial set of around fifty, which allow to obtain a good predictive performance (R-2=0.79). These variables reveal crucial aspects for the definition of management strategies focused on promoting academic success.

Abstract
With the aim of disseminating the potential and the capacity of Educational Data Mining (EDM) as an instrument of investigation and analysis in the support to the management of Higher Education Institutions, this paper presents a brief description of some of the most relevant studies in the area. The analysis carried out allows to highlight the innovations that EDM has been promoting, as well as current and future research trends.

Abstract
The early classification of university students according to their potential academic performance can be a useful strategy to mitigate failure, to promote the achievement of better results and to better manage resources in higher education institutions. This paper proposes a two-stage model, supported by data mining techniques, that uses the information available at the end of the first year of students' academic career (path) to predict their overall academic performance. Unlike most literature on educational data mining, academic success is inferred from both the average grade achieved and the time taken to conclude the degree. Furthermore, this study proposes to segment students based on the dichotomy between the evidence of failure or high performance at the beginning of the degree program, and the students' performance levels predicted by the model. A data set of 2459 students, spanning the years from 2003 to 2015, from a European Engineering School of a public research University, is used to validate the proposed methodology. The empirical results demonstrate the ability of the proposed model to predict the students' performance level with an accuracy above 95%, in an early stage of the students' academic path. It is found that random forests are superior to the other classification techniques that were considered (decision trees, support vector machines, naive Bayes, bagged trees and boosted trees). Together with the prediction model, the suggested segmentation framework represents a useful tool to delineate the optimum strategies to apply, in order to promote higher performance levels and mitigate academic failure, overall increasing the quality of the academic experience provided by a higher education institution.