Abstract
Assuring the quality control of publications in the scientific literature is one of the main challenges of the peer review process. Consequently, there has been an increasing demand for computing solutions that will help to maintain the quality of this process. Recently, the use of Artificial Intelligence techniques has been highlighted, applied in the detection of plagiarism, bias, among other functions. The assessment of the reviewer's review has also been considered as important in the process, but, little is known about it, for instance, which techniques have been applied in this assessment or which criteria have been assessed. Therefore, this systematic literature review aims to find evidence regarding the computational approaches that have been used to evaluate reviewers' reports. In order to achieve this, five online databases were selected, from which 72 articles were identified that ma the inclusion criteria of this review, all of which have been published since 2000. The result returned 10 relevant studies meeting the evaluation requirements of scientific article reviews. The review revealed that mechanisms to rank review reports according to a score, as well as the word analysis, are the most common tools, and that there is no consensus on quality criteria. The systematic literature review has shown that reviewers' report assessment is a valid tool for maintaining quality throughout the process. However, it still needs to be further developed if it is to be used as a resource which surpass a single conference or journal, making the peer review process more rigorous and less based on random choice.

Abstract
Most modern solutions for video game balancing are directed towards specific games. We are currently researching general methods for automatic multiplayer game balancing. The problem is modeled as a meta-game, where game-play change the rules from another game. This way, a Machine Learning agent that learns to play a meta-game, learns how to change a base game following some balancing metric. But an issue resides in the generation of high volume of game-play training data, was agents of different skill compete against each other. For this end we propose the automatic generation of a population of surrogate agents by learning sampling. In Reinforcement Learning an agent learns in a trial error fashion where it improves gradually its policy, the mapping from world state to action to perform. This means that in each successful evolutionary step an agent follows a sub-optimal strategy, or eventually the optimal strategy. We store the agent policy at the end of each training episode. The process is evaluated in simple environments with distinct properties. Quality of the generated population is evaluated by the diversity of the difficulty the agents have in solving their tasks. © Springer Nature Switzerland AG 2019.

Abstract

Abstract

Abstract
In recent years, public institutions have undergone a progressive modernization process, bringing several administrative services to be provided electronically. Some institutions are responsible for analyzing citizen complaints, which come in huge numbers and are mainly provided in free-form text, demanding for some automatic way to process them, at least to some extent. In this work, we focus on the task of automatically identifying economic activities in complaints submitted to the Portuguese Economic and Food Safety Authority (ASAE), employing natural language processing (NLP) and machine learning (ML) techniques for Portuguese, which is a language with few resources. We formulate the task as several multi-class classification problems, taking into account the economic activity taxonomy used by ASAE. We employ features at the lexical, syntactic and semantic level using different ML algorithms. We report the results obtained to address this task and present a detailed analysis of the features that impact the performance of the system. Our best setting obtains an accuracy of 0.8164 using SVM. When looking at the three most probable classes according to the classifier’s prediction, we report an accuracy of 0.9474. © 2019, Springer Nature Switzerland AG.

Abstract