Abstract
Dental implant recognition in patients without available records is a time-consuming and not straightforward task. The traditional method is a complete user-dependent process, where the expert compares a 2D X-ray image of the dental implant with a generic database. Due to the high number of implants available and the similarity between them, automatic/semi-automatic frameworks to aide implant model detection are essential. In this study, a novel computer-aided framework for dental implant recognition is suggested. The proposed method relies on image processing concepts, namely: (i) a segmentation strategy for semi-automatic implant delineation; and (ii) a machine learning approach for implant model recognition. Although the segmentation technique is the main focus of the current study, preliminary details of the machine learning approach are also reported. Two different scenarios are used to validate the framework: (1) comparison of the semi-automatic contours against implant's manual contours of 125 X-ray images; and (2) classification of 11 known implants using a large reference database of 601 implants. Regarding experiment 1, 0.97+/-0.01, 2.24+/-0.85 pixels and 11.12+/-6 pixels of dice metric, mean absolute distance and Hausdorff distance were obtained, respectively. In experiment 2, 91% of the implants were successfully recognized while reducing the reference database to 5% of its original size. Overall, the segmentation technique achieved accurate implant contours. Although the preliminary classification results prove the concept of the current work, more features and an extended database should be used in a future work.

Abstract
Developments in advanced autonomous production resources have increased the interest in the Single-Machine Scheduling Problem (SMSP). Until now, researchers used SMSP with little to no practical application in industry, but with the introduction of multi-purpose machines, able of executing an entire task, such as 3D Printers, replacing extensive production chains, single-machine problems are becoming a central point of interest in real-world scheduling. In this paper we study how simple, easy to implement, Just-in-Time (JIT) based, constructive heuristics, can be used to optimize customer and enterprise oriented performance measures. Customer oriented performance measures are mainly related to the accomplishment of due dates while enterprise-oriented ones typically consider other time-oriented measures.

Abstract
Optimization in current decision support systems has a highly interdisciplinary nature related with the need to integrate different techniques and paradigms for solving real-world complex problems. Computing optimal solutions in many of these problems are unmanageable. Heuristic search methods are known to obtain good results in an acceptable time interval. However, parameters need to be adjusted to allow good results. In this sense, learning strategies can enhance the performance of a system, providing it with the ability to learn, for instance, the most suitable optimization technique for solving a particular class of problems, or the most suitable parameterization of a given algorithm on a given scenario. Hyper-heuristics arise in this context as efficient methodologies for selecting or generating (meta) heuristics to solve NP-hard optimization problems. This paper presents the specification of a hyper-heuristic for selecting techniques inspired in nature, for solving the problem of scheduling in manufacturing systems, based on previous experience. The proposed hyper-heuristic module uses a reinforcement learning algorithm, which enables the system with the ability to autonomously select the meta-heuristic to use in optimization process as well as the respective parameters. A computational study was carried out to evaluate the influence of the hyper-heuristics on the performance of a scheduling system. The obtained results allow to conclude about the effectiveness of the proposed approach.

Abstract
Metaheuristics are very useful to achieve good solutions in reasonable execution times. Sometimes they even obtain optimal solutions. However, to achieve near-optimal solutions, the appropriate tuning of parameters is required. This paper presents a Racing based learning module proposal for an autonomous parameter tuning of Metaheuristics. After a literature review on Metaheuristics parameter tuning and Racing approaches, the learning module is presented. A computational study for the resolution of the Scheduling problem is also presented. Comparing the preliminary obtained results with previous published results allow to conclude about the effectiveness and efficiency of this proposal. © 2015 AISTI.

Abstract
Manufacturing environments require a real-time adaptation and optimization method to dynamically and intelligently maintain the current scheduling plan feasible. This way, the organization keeps clients satisfied and achieves its objectives (costs are minimized and profits maximized). This paper proposes an optimization approach - Selection Constructive based Hyper-heuristic for Dynamic Scheduling - to deal with these dynamic events, with the main goal of maintaining the current scheduling plan feasible and robust as possible. The development of this dynamic adaptation approach is inspired on evolutionary computation and hyper-heuristics. Our empirical results show that a selection constructive hyper-heuristic could be advantageous on solving dynamic adaptation optimization problems. © 2015 AISTI.

Abstract
User modelling has become a central subject for anybody interested in understanding how users interact with technology. Personalization is a key issue in an era when there is so much information and so many people interacting in so many ways. Modern users desire a customized experience that adapts itself to their requirements and understands what they need even before they notice it. In order to morph any system into an adapting one, every relevant interaction with its users has to be maintained. Then, a mathematical structure capable of discovering patterns amongst that information is necessary, being able to classify users according to the roles they play. With a correct user categorization, the system knows when, how and what to do to adapt its content, via a mixed-initiative approach. In this paper, an artificial neural network is selected as classifier and users are divided in three roles, from beginners to experts. ADSyS, the target system of this proposal, adapts its content based on who is operating it, providing a higher usability. This guide on how to adapt a system to its users is built as part of ADSyS, but is intended to be generalized as a foundation to other systems. © 2015 AISTI.