Abstract
The information provided by online traveler reviews is becoming a key element in the decision-making process of hotel customers, reducing the uncertainty and the perceived risk of a traveler. Therefore, a careful analysis of the content provided by online customers' reviews might give invaluable information concerning the key determinants, from a user's perspective, of the quality of the service provided, justifying the attributed service rating. The objectives of this study are twofold: (1) use text-mining techniques to analyze the user's generated content automatically collected from hotels in Porto in a certain period of time and, from this analysis, derive the most frequent terms used to describe the service; (2) understand whether it is possible to predict the aggregated rating assigned by reviewers based on the terms used and, at the same time, identify the terms showing high predictive capacity. Our study attempts to support hotel service managers in achieving their strategic and tactical goals by using innovative text- and data-mining tools to explore the wealth of information provided by user generated content in an easy and timely way.

Abstract
Currently, many libraries, either academic or public, possess information systems to support their operations. Although libraries are becoming more aware of the potential of data analytics in supporting library management decisions, there is still a long way to go to take plenty advantage of the information collected. This paper proposes a prescriptive analytics solution to enhance the service provided by libraries, by optimizing libraries layout. The quantitative method introduced aims to identify layout configurations that minimize the time spent by clients in picking books from the library. A new multi-floor layout optimization algorithm is developed, based on the pairwise exchange method heuristic. A real data sample of approximately 66.000 loans, taken from the information system of a European Engineering School's library, was analyzed and processed. The method proposed was used to improve the library's current departments configuration, achieving an improvement of 13.2% in terms of walking distance to collect the books. The results corroborate the effectiveness of the method proposed and its potential in supporting library management decisions.

Abstract
This paper describes a study on applying data mining techniques to power transformer failure prediction. The data set used consisted not only on DGA tests, but also in other tests done to the transformer’s insulating oil. This dataset presented several challenges, such as highly imbalanced classes (common in failure prediction problems), and the temporal nature of the observations. To overcome these challenges, several techniques were applied for prediction and better understand the dataset. Pre-processing and temporality incorporation in the dataset is discussed. For prediction, a 1-class and 2-class SVM, decision trees and random forests, as well as a LSTM neural network were applied to the dataset. As the prediction performance was low (high false-positive rate), we conducted a test to ascertain if the amount of data collected was sufficient. Results indicate that the frequency of data collection was not adequate, hinting that the degradation period was shorter than the periodicity of data collection.

Abstract

Abstract
Assessing eco-efficiency of companies is important to ensure the creation of wealth without compromising the needs of future generations. This work aims to extend the eco-efficiency concept by including in the assessment new features related to environmental benefits and environmental burdens. This concept is implemented using an innovative Directional Distance Function model, which searches for improvements in the magnitude of the indicators and in the composition of the resources consumed. This framework can help firms to become more sustainable by replacing non-renewable inputs with "greener" alternatives. We present an empirical application to large mining companies. Different scenarios regarding managerial priorities for adjustments to firms' economic and environmental indicators are explored. The results obtained and their managerial implications are discussed in the context of mining firms activity.

Abstract
130In the majority of European countries, the evaluation of schools is at the heart of the educational system as a means to guarantee the quality of education. Every year, in most countries around the world, students perform national exams. Their results are analyzed by several stakeholders, including governmental agencies, the media, and researchers on educational issues. At present, advances in information and communication technology (ICT) and data analysis techniques allow schools to make use of massive amounts of data in their daily management. This chapter focuses in particular on the use of students’? data to benchmark schools. It illustrates the potential contribution of the information gathered and analyzed through data analytics to promote the continuous improvement of schools’? educational processes. © 2018 by Taylor & Francis Group, LLC.