Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu
Publications

2018

Merging Datasets for Hate Speech Classification in Italian

Authors
Fortuna, P; Bonavita, I; Nunes, S;

Publication
Proceedings of the Sixth Evaluation Campaign of Natural Language Processing and Speech Tools for Italian. Final Workshop (EVALITA 2018) co-located with the Fifth Italian Conference on Computational Linguistics (CLiC-it 2018), Turin, Italy, December 12-13, 2018.

Abstract
This paper presents an approach to the shared task HaSpeeDe within Evalita 2018. We followed a standard machine learning procedure with training, validation, and testing phases. We considered word embedding as features and deep learning for classification. We tested the effect of merging two datasets in the classification of messages from Facebook and Twitter. We concluded that using data for training and testing from the same social network was a requirement to achieve a good performance. Moreover, adding data from a different social network allowed to improve the results, indicating that more generalized models can be an advantage.

2018

Layered logics, coalgebraically

Authors
Barbosa, LS;

Publication
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Abstract
This note revisits layered logics from a coalgebraic point of view, and proposes a naturality condition to express the typical hierarchical requirement under which all abstract transitions should be traceable in more specialised layers. © Springer International Publishing AG 2018.

2018

An Outline of Fault-Tolerant Control System for Electric Vehicles Operating in a Platoon

Authors
Lopes, A; Araujo, RE;

Publication
TECHNOLOGICAL INNOVATION FOR RESILIENT SYSTEMS (DOCEIS 2018)

Abstract
High level vehicle automation systems are currently being studied to attenuate highway traffic and energy consumption by applying the concept of platooning, which has gained increased attention due to progresses in the next generation of mobile communication (5G). The introduction of more complex automation systems originates, however, fault entry points that hinders the system safety and resilience. This paper presents an initial control architecture for the electric vehicles platoon from a fault-tolerant control perspective. To achieve a fault-tolerant platooning structure an over-actuated electric vehicle topology is proposed which may allow the implementation of different redundancies. Furthermore, some of the major challenges in the platooning network control system (NCS) are presented and the techniques to overcome these issues are explored.

2018

Integrating pricing and capacity decisions in car rental: A matheuristic approach

Authors
Oliveira, BB; Carravilla, MA; Oliveira, JF;

Publication
OPERATIONS RESEARCH PERSPECTIVES

Abstract
Pricing and capacity decisions in car rental companies are characterized by high flexibility and interdependence. When planning a selling season, tackling these two types of decisions in an integrated way has a significant impact. This paper tackles the integration of capacity and pricing problems for car rental companies. These problems include decisions on fleet size and mix, acquisitions and removals, fleet deployment and repositioning, as well as pricing strategies for the different rental requests. A novel mathematical model is proposed, which considers the specific dynamics of rentals on the relationship between inventory and pricing as well as realistic requirements from the flexible car rental business, such as upgrades. Moreover, a solution procedure that is able to solve real-sized instances within a reasonable time frame is developed. The solution procedure is a matheuristic based on the decomposition of the model, guided by a biased random-key genetic algorithm (BRKGA) boosted by heuristically generated initial solutions. The positive impact on profit, of integrating capacity and pricing decisions versus a hierarchical/sequential approach, is validated.

2018

Stability Analysis of a Synchronous Generator-Based Control Technique used in Large-Scale Grid Integration of Renewable Energy

Authors
Mehrasa, M; Sepehr, A; Pouresmaeil, E; Kyyra, J; Marzband, M; Catalao, JPS;

Publication
2018 INTERNATIONAL CONFERENCE ON SMART ENERGY SYSTEMS AND TECHNOLOGIES (SEST)

Abstract
A synchronous generator (SG)-based control technique is proposed in this paper to force the grid voltage magnitude and frequency to follow the desired values under high penetration of renewable energy sources. A more detailed relationship between power dynamic model and SGs characteristics has been considered in the proposed control strategy. The active and reactive power error-based curve of the proposed control technique is evaluated in detail. Besides, the grid angular frequency error based on the proposed control technique performance is assessed in the next step. Simulation is employed in Matlab/Simulink to verify the operation of the proposed control technique in power converters for large-scale integration of renewable energy sources into the power grids.

2018

MULTI-TEMPORAL INSAR MONITORING OF THE ASWAN HIGH DAM (EGYPT)

Authors
Ruiz Armenteros, AM; Delgado, JM; Lamas Fernandez, F; Bravo Pareja, R; Lazecky, M; Bakon, M; Sousa, JJ; Caro Cuenca, M; Verstraeten, G; Hanssen, RF;

Publication
IGARSS 2018 - 2018 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM

Abstract
The Aswan High Dam, Egypt, was built in the 1960s and is one of the biggest dams in the world. It stopped the seasonal flood of Nile river allowing the urban expansion of cities/villages and the full year cultivation, producing 10x10(9) kWh of power annually. The dam is located in an area where several earthquakes (M-L<6) occurred from 1981 to 2007. In this paper, we want to identify any potential damage that could be caused to the dam, and assess its overall structural stability using Multi-Temporal InSAR (MT-InSAR). To reach this goal, we process Envisat data from descending orbits acquired between 2003 and 2010. Our initial estimates show relatively small rates (maximum around -3 mm/yr in the satellite Line-Of-Sight) of subsidence, whose implications must be further investigated. In addition, we perform a preliminary stress-strain analysis of the dam using FEL and FEM methods to assess if the detected movements correspond to the expected vertical behavior for such mega-structure.

  • 1693
  • 4186