2020
Autores
Carneiro, D; Guimarães, M; Sousa, M;
Publicação
Hybrid Intelligent Systems - 20th International Conference on Hybrid Intelligent Systems (HIS 2020), Virtual Event, India, December 14-16, 2020
Abstract
Machine Learning systems are generally thought of as fully automatic. However, in recent years, interactive systems in which Human experts actively contribute towards the learning process have shown improved performance when compared to fully automated ones. This may be so in scenarios of Big Data, scenarios in which the input is a data stream, or when there is concept drift. In this paper we present a system for supporting auditors in the task of financial fraud detection. The system is interactive in the sense that the auditors can provide feedback regarding the instances of the data they use, or even suggest new variables. This feedback is incorporated into newly trained Machine Learning models which improve over time. In this paper we show that the order by which instances are evaluated by the auditors, and their feedback incorporated, influences the evolution of the performance of the system over time. The goal of this paper is to study of different instance selection strategies for Human evaluation and feedback can improve the learning speed. This information can then be used by the system to determine, at each moment, which instances would improve the system the most, so that these can be suggested to the users for validation. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.
2020
Autores
Ramos, D; Carneiro, D; Novais, P;
Publicação
INTELLIGENT DISTRIBUTED COMPUTING XIII
Abstract
Machine Learning is a field in which significant steps forward have been taken in the last years, resulting in a wide variety of available algorithms, for many different problems. Nonetheless, most of these algorithms focus on the training of static models, in the sense that the model stops evolving after the training phase. This is increasingly becoming a limitation, especially in an era in which datasets are increasingly larger and may even arrive as sequential streams of data. Frequently retraining a model, in these scenarios, is not realistic. In this paper we propose evoRF: a combination of a Random Forest with an evolutionary approach. Its key innovative aspect is the evolution of the weights of the Random Forest over time, as new data arrives, thus making the forest's voting scheme adapt to the new data. Older trees can also be replaced by newly trained ones, according to their accuracy, ensuring that the ensemble remains up to date without requiring a whole retraining.
2020
Autores
Carneiro, D; Silva, F; Guimarães, M; Sousa, D; Novais, P;
Publicação
Ambient Intelligence - Software and Applications - 11th International Symposium on Ambient Intelligence, ISAmI 2020, L'Aquila, Italy, October 7 - 9, 2020
Abstract
The main focus of an Intelligent environment, as with other applications of Artificial Intelligence, is generally on the provision of good decisions towards the management of the environment or the support of human decision-making processes. The quality of the system is often measured in terms of accuracy or other performance metrics, calculated on labeled data. Other equally important aspects are usually disregarded, such as the ability to produce an intelligible explanation for the user of the environment. That is, asides from proposing an action, prediction, or decision, the system should also propose an explanation that would allow the user to understand the rationale behind the output. This is becoming increasingly important in a time in which algorithms gain increasing importance in our lives and start to take decisions that significantly impact them. So much so that the EU recently regulated on the issue of a “right to explanation”. In this paper we propose a Human-centric intelligent environment that takes into consideration the domain of the problem and the mental model of the Human expert, to provide intelligible explanations that can improve the efficiency and quality of the decision-making processes. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.
2020
Autores
Ramos, D; Carneiro, D; Novais, P;
Publicação
AI COMMUNICATIONS
Abstract
The requirements of Machine Learning applications are changing rapidly. Machine Learning models need to deal with increasing volumes of data, and need to do so quicker as responses are expected more than ever in real-time. Plus, sources of data are becoming more and more dynamic, with patterns that change more frequently. This calls for new approaches and algorithms, that are able to efficiently deal with these challenges. In this paper we propose the use of a Genetic Algorithm to Optimize a Stacking Ensemble specifically developed for streaming scenarios. A pool of solutions is maintained in which each solution represents a distribution of weights in the ensemble. The Genetic Algorithm continuously optimizes these weights to minimize the cost function. Moreover, new models are added at regular intervals, trained on more recent data. These models eventually replace older and less accurate ones, making the ensemble adapt continuously do changes in the distribution of the data.
2020
Autores
Teixeira, A; Rodrigues, M; Carneiro, D; Novais, P;
Publicação
INTELLIGENT SYSTEMS AND APPLICATIONS, VOL 1
Abstract
Emotion is an essential part of what means to be human, but it is still disregarded by most technical fields as something not to be considered in scientific or engineering projects. However, the understanding of emotion as an aspect of decision-making processes and of modelling of human behavior is essential to create a better connection between humans and their tools and machines. With this work we focus on the measurement of emotion of users through the use of non-intrusive methods, like measuring inputs and reactions to stimuli, along with the creation of a tool that measures the emotional changes caused by visual output created by the tool itself. Usage of the tool in a test environment and the subsequent analysis of the data obtained will allow for conclusions about the effectiveness of the method, and if it is possible to apply it to future studies on human emotions by investigators in the fields of psychology and computation.
2020
Autores
Carneiro, D; Veloso, P; Ventura, A; Palumbo, G; Costa, J;
Publicação
Intelligent Data Engineering and Automated Learning - IDEAL 2020 - 21st International Conference, Guimaraes, Portugal, November 4-6, 2020, Proceedings, Part II
Abstract
As technology evolves, frauds, in all their different forms, become increasingly more complex, as mega-processes such as Luanda Leaks or Panama Papers have recently shown. Fraud detection mechanisms must thus also resort to recent technological developments to ensure that even new forms of complex fraud are detected in time. In this paper we present a system to facilitate fraud detection in Public Procurement in Portugal. It relies on three main components. Data on public procurement and involved entities is acquired from public sources and integrated into a graph-oriented database. A rules-engine enriches these data with information additional information, using legal rules or custom rules defined by the users. Finally, a graph-oriented User Interface is used to support decision-making, allowing users to quickly and efficiently explore and filter information, in a natural and geo-referenced way. The main goal of this system is to increase transparency by facilitating access to relevant information, and in this way contribute to the fairness of the whole public procurement procedure. © 2020, Springer Nature Switzerland AG.
The access to the final selection minute is only available to applicants.
Please check the confirmation e-mail of your application to obtain the access code.