Abstract
Current challenges in machine learning include dealing with temporal data streams, drift and non-stationary scenarios, often with text data, whether in social networks or in business systems. This dynamic nature tends to limit the performance of traditional static learning models and dynamic learning strategies must be put forward. However, acquiring the performance of those strategies is not a straightforward issue, as sample's dependency undermines the use of validation techniques, like crossvalidation. In this paper we propose to use the McNemar's test to compare two distinct approaches that tackle adaptive learning in dynamic environments, namely DARK (Drift Adaptive Retain Knowledge) and Learn++. NSE (Learn++ for Non-Stationary Environments). The validation is based on a Twitter case study benchmark constructed using the DOTS (Drift Oriented Tool System) dataset generator. The results obtained demonstrate the usefulness and adequacy of using McNemar's statistical test in dynamic environments where time is crucial for the learning algorithm. © 2018 IEEE.

Abstract

Abstract
Information and Communication Technologies appear one of the most critical areas to the success of tourism in the future and the way it will promote the tourist destinations. This paper aims to analyze the relevance given by the various Portuguese institutions of higher education to Information and Communication Technologies in their degrees. The analysis carried out was done in degree courses operating in this academic year, 2017/2018, in Portuguese universities and polytechnics, public and private. A comparison was also made with the reality of 2012/2013, since as undergraduate degrees have been evaluated by the Agency for Assessment and Accreditation of Higher Education (A3ES), often making changes, resulting from the evaluation.

Abstract

Abstract
Olive oil and wine production have a worldwide economic impact. Their market reliability is under great concern because of the increasing number of fraud and adulteration attempts. The need for a traceability system in all its extension is crucial particularly for the cases of olive oils and wines with certified labels, in which only a limited number of olives and grapevine varieties, respectively, are allowed in a restricted well-defined geographical area. Molecular markers have been vastly applied to the food sector, and in particular High-Resolution DNA Melting technology has been successfully applied for olive oil and wine authentication, as part of the traceability system. In this review, the applications of HRM and their usefulness for this sector considering, Safety, Security and Authenticity will be reviewed. A broad overview of the HRM technique will be presented, focusing on the aspects that are crucial for its success, in particular the new generation of fluorescent dsDNA dyes used for amplicon detection and quantification, and the data analysis. A brief outlook on the olive oil and wine authenticity procedures, based on new DNA technology advances, and in which way this may influence the future establishment of a traceability system will be discussed.

Abstract
Extraction and normalization of temporal expressions are essential for many NLP tasks. While a considerable effort has been put on this task over the last few years, most of the research has been conducted on the English domain, and only a few works have been developed on other languages. In this paper, we present ParsTime, a tagger for temporal expressions in Persian (Farsi) documents. ParsTime is a rule-based system that extracts and normalizes Persian temporal expressions according to the TIMEX3 annotation standard. Our experimental results show that ParsTime can identify temporal expressions in Persian texts with an F1-score 0.89. As an additional contribution we make available our code to the research community.