Abstract
Knowledge Discovery in Databases (KDD) is a major innovation in knowledge extraction. This knowledge can be extracted to recognize patterns or behaviors. Board games playing patterns are a concise experiment on testing data mining methods in order to find such patterns and behaviors. In this work a Connect-4 game is simulated with several distinct players with different characteristics. Most of these distinct players have intelligent game playing abilities, whereas others are simpler and play by very simple rules. The work uses three different data-mining algorithms in order to classify the players and their moves. Analyzing the results achieved we can conclude that General Linear Model leads to better results in terms of accuracy, class precision and class recall

Abstract
With the growing interest in research in poker, by scientists belonging to the area of Artificial Intelligence, has arisen the need to overcome the imperfect information of the same, as a stochastic game, by the challenges that this enables. In this work we intend to create data models that allow us to know the plays of a real player, supporting the decision to play in the pre-flop phase, within the Texas Hold'em side. To accomplish this, several Data Mining techniques were applied for classification, using RapidMiner software. The development of this work was based on data from online games of professional poker players, who had important information of the factors inherent to a real game.

Abstract
This work addresses the field of the artificial devices that replace a missing body limb, particularly the prosthesis used upon a transtibial (TT) amputation, also known as ankle-foot prosthesis. A major challenge in designing an ankle-foot prosthesis lies in synthesizing a mechanism that will functionally mimic the missing part of an anatomical leg and, therefore, restoring the normal gait of the amputee. This way, it will be discussed and presented the requirements for a TT prosthesis, the steps that lead to the primary sketches of the prosthetic model and its main mechanism. This work constitutes a preliminary research in the field of the bionic TT prosthesis.

Abstract
The national panel for higher education is a big social impact event, one which mobilizes thousands of candidates. However, the heterogeneity of the Portuguese university and polytechnic infrastructure and the sheer dimension of the reality in study makes an eventual interpretation of the data obtained from that panel, and the official data only present generic and global information. This work will bring to light information with added value to those responsible on these institutions, in their decision taking processes by extracting data from the education minister site and processing it using data mining techniques.

Abstract
The continuous monitoring of human gait would allow to more objectively verify the abnormalities that arise from the most common pathologies. Therefore, this manuscript proposes a real-time tool for human gait detection from lower trunk acceleration. The vertical acceleration signal was acquired through an IMU mounted on a waistband, a wearable device. The proposed algorithm was based on a finite state machine (FSM) which includes a set of suitable decision rules and the detection of Heel-Strike (HS), Foot-flat (FF), Toe-off (TO), Mid-Stance (MS) and Heel-strike (HS) events for each leg. Results involved 7 healthy subjects which had to walk 20 m three times with a comfortable speed. The results showed that the proposed algorithm detects in real-time all the mentioned events with a high accuracy and time-effectiveness character. Also, the adaptability of the algorithm has also been verified, being easily adapted to some gait conditions, such as for different speeds and slopes. Further, the developed tool is modular and therefore can easily be integrated in another robotic control system for gait rehabilitation. These findings suggest that the proposed tool is suitable for the real-time gait analysis in real-life activities. © Springer International Publishing AG, part of Springer Nature 2018.

Abstract
The Airline Operations Control Center (AOCC) has the responsibility to ensure that flights meet their planned schedule or, if any problem arises, to find a viable solution that minimizes both the impact in the operational plan and its cost. The high cost of resources involved in this process (aircraft and crew members) leads to a lack of additional resources from the airline companies, implying a restricted solution space. Here, we propose an electronic market modeled as a multi-agent system where airline companies can negotiate and lease each other the required resources when solving a disruption problem, thus expanding their solution space. The proposed negotiation occurs in several rounds, where qualitative comments made by the buyer agent on proposals sent by the sellers enables these to learn how to calculate new proposals, using a case-based reasoning methodology.