8 results on '"Baptista, Cláudio"'
Search Results
2. Using Machine Learning and NLP for the Product Matching Problem
- Author
-
de Santana, Matheus Alcantara, de Souza Baptista, Cláudio, Alves, André Luiz Firmino, Firmino, Anderson Almeida, da Silva Januário, Gerson, da Silva Caldera, Roney Wellington, Kacprzyk, Janusz, Series Editor, Gomide, Fernando, Advisory Editor, Kaynak, Okyay, Advisory Editor, Liu, Derong, Advisory Editor, Pedrycz, Witold, Advisory Editor, Polycarpou, Marios M., Advisory Editor, Rudas, Imre J., Advisory Editor, Wang, Jun, Advisory Editor, Nagar, Atulya K., editor, Singh Jat, Dharm, editor, Mishra, Durgesh Kumar, editor, and Joshi, Amit, editor
- Published
- 2023
- Full Text
- View/download PDF
3. COVID-19 Diagnosis Through Deep Learning Techniques and Chest X-Ray Images
- Author
-
Negreiros, Ramoni Reus Barros, Silva, Isabel Heloíse Santos, Alves, André Luiz Firmino, Valadares, Dalton Cézane Gomes, Perkusich, Angelo, and Baptista, Cláudio de Souza
- Published
- 2023
- Full Text
- View/download PDF
4. Using Machine Learning for Risk Classification in Brazilian Federal Voluntary Transfers
- Author
-
Guilhon, Daniel M., de Oliveira, Aillkeen Bezerra, Gomes, Daniel L., Jr, Paiva, Anselmo C., de Souza Baptista, Cláudio, Junior, Geraldo Braz, de Almeida, João Dallysson Sousa, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Kö, Andrea, editor, Francesconi, Enrico, editor, Kotsis, Gabriele, editor, Tjoa, A Min, editor, and Khalil, Ismail, editor
- Published
- 2021
- Full Text
- View/download PDF
5. A Machine Learning Approach for Classifying Road Accident Hotspots.
- Author
-
Amorim, Brunna de Sousa Pereira, Firmino, Anderson Almeida, Baptista, Cláudio de Souza, Júnior, Geraldo Braz, Paiva, Anselmo Cardoso de, and Júnior, Francisco Edeverton de Almeida
- Subjects
MACHINE learning ,TRAFFIC accidents ,RANDOM forest algorithms ,SUPERVISED learning ,FEATURE selection ,WEATHER - Abstract
Road accidents are a worldwide problem, affecting millions of people annually. One way to reduce such accidents is to predict risk areas and alert drivers. Advanced research has been carried out on identifying accident-influencing factors and potential highway risk areas to mitigate the number of road accidents. Machine learning techniques have been used to build prediction models using a supervised classification based on a labeled dataset. In this work, we experimented with many machine learning algorithms to discover the best classifier for the Brazilian federal road hotspots associated with severe or nonsevere accident risk using several features. We tested with SVM, random forest, and a multi-layer perceptron neural network. The dataset contains a ten-year road accident report by the Brazilian Federal Highway Police. The feature set includes spatial footprint, weekday and time when the accident happened, road type, route, orientation, weather conditions, and accident type. The results were promising, and the neural network model provided the best results, achieving an accuracy of 83%, a precision of 84%, a recall of 83%, and an F1-score of 82%. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
6. Semi-automatic photograph tagging by combining context with content-based information.
- Author
-
de Figueirêdo, Hugo Feitosa, de Souza Baptista, Cláudio, Casanova, Marco Antonio, da Silva, Tiago Eduardo, and de Paiva, Anselmo Cardoso
- Subjects
- *
PHOTOGRAPHS , *INFORMATION theory , *HUMAN facial recognition software , *MACHINE learning , *REGRESSION analysis - Abstract
This article proposes a semi-automatic technique for the annotation of people in photographs. The technique uses context and content information and is based on a weighted sum of estimators, which results in a list of the person’s contacts that are more likely to be present in a photograph. Machine learning methods, such as multivariable linear regression and slope function, are adopted to filter and weight the estimators and eigenfaces for face recognition. The article also describes the results of experiments that were performed with a collection of 4050 photographs with 365 different people, which indicate that the proposed technique outperforms techniques that adopt only context or only content using as a performance metric the H-Hit rate of correct annotations. [ABSTRACT FROM AUTHOR]
- Published
- 2015
- Full Text
- View/download PDF
7. Risk classification in federal voluntary transfers using XGBoost
- Author
-
GUILHON, Daniel Moreira, PAIVA, Anselmo Cardoso, GOMES JÚNIOR, Daniel Lima, BRAZ JÚNIOR, Geraldo, and BAPTISTA, Cláudio de Souza
- Subjects
Predição de risco ,Machine Learning ,Transferências voluntárias ,Ciência da Computação ,Aprendizagem computacional ,Risk Prediction ,XGBoost ,Voluntary Transfers - Abstract
Submitted by Daniella Santos (daniella.santos@ufma.br) on 2021-01-27T14:27:32Z No. of bitstreams: 1 DanielGuilhon.pdf: 3336701 bytes, checksum: 442d8c723041ca36ea4dfad7c7adfa7d (MD5) Made available in DSpace on 2021-01-27T14:27:32Z (GMT). No. of bitstreams: 1 DanielGuilhon.pdf: 3336701 bytes, checksum: 442d8c723041ca36ea4dfad7c7adfa7d (MD5) Previous issue date: 2020-07-16 After the Brazilian re-democratization, states and municipalities had to rely on federal government’s voluntary transfers of resources to achieve their public policies. For greater timeliness in the recovery of resources that may have been spent inappropriately, it is necessary to assign risk profiles of success or failure of these transfers. In this work, we propose a methodology that uses eXtreme Gradient Boosting (XGBoost) algorithm, using balanced and unbalanced data sets, with the use of hyperparameter optimization techniques, such as Tree-structured Parzen Bayesian Estimator (TPE). The results achieved good success rates. Results for XGBoost using balanced data showed a recall of 89.3% and unbalanced data a recall of 87.8%. However, for unbalanced data, the AUC score was 98.1%, against 97.9% for balanced data. Incorporating information data about the agreed object using natural language processing techniques can improve the results obtained. Com a redemocratização no Brasil, estados e municípios passaram a contar com transferências voluntárias de recursos por parte do Governo Federal para a consecução de suas políticas públicas. Para uma maior tempestividade na recuperação de recursos eventualmente gastos de forma inadequada, é necessária uma ferramenta de classificação para atribuir perfis de risco de sucesso ou fracasso dessas transferências. Neste trabalho, propomos o uso do algoritmo eXtreme Gradient Boosting (XGBoost) usando conjuntos de dados balanceados e desbalanceados, com técnicas de otimização de hiperparâmetros Tree-structured Parzen Estimator bayesiano (TPE). Os resultados alcançaram boas taxas de sucesso. Os resultados do XGBoost mostraram uma taxa de sensibilidade usando dados balanceados de 89,3% e dados desbalanceados 87,8%. No entanto, para os dados desbalanceados, a AUC foi de 98,1%, contra 97,9% para os dados balanceados. Incorporar dados como informações acerca do objeto pactuado utilizando-se técnicas de processamento de linguagem natural pode melhorar os resultados obtidos.
- Published
- 2020
8. Use of machine learning for classification risk of road accidents
- Author
-
AMORIM, Brunna de Sousa Pereira., BAPTISTA, Cláudio de Souza., GOMES , Herman Martins., and BRAZ JUNIOR, Geraldo.
- Subjects
Machine Learning ,Automated Machine Learning ,Aprendizado de Máquina ,Risk Rating ,Ciência da Computação ,Redução de Dimensionalidade ,Aprendizado de Máquina Automatizado ,Seleção de Características ,Risco de Acidente em Rodovias ,Road Accident Risk ,Feature Selection ,Classificação de Risco ,Dimensionality Reduction - Abstract
Submitted by Maria Medeiros (maria.dilva1@ufcg.edu.br) on 2019-09-26T11:30:54Z No. of bitstreams: 1 BRUNNA DE SOUSA PEREIRA AMORIM - DISSERTAÇÃO (PPGCC) 2019.pdf: 838135 bytes, checksum: 1d867ff60a16840dcdd92bfda964ba5a (MD5) Made available in DSpace on 2019-09-26T11:30:54Z (GMT). No. of bitstreams: 1 BRUNNA DE SOUSA PEREIRA AMORIM - DISSERTAÇÃO (PPGCC) 2019.pdf: 838135 bytes, checksum: 1d867ff60a16840dcdd92bfda964ba5a (MD5) Previous issue date: 2019-08-21 CNPq Soluções para identificação dos fatores que influenciam o acontecimento de acidentes em rodovias e a identificação de trechos de risco estão sendo estudados e aplicados por pesquisadores e governos de todo o mundo, a fim de encontrar uma solução que possa diminuir o número de tais acidentes. No entanto, o estudo de acidentes em rodovias depende do local onde o mesmo acontece. Destarte, esta pesquisa faz uso de técnicas de aprendizado de máquina supervisionado e aprendizado de máquina automatizado com o uso de diferentes características para analisar seu impacto na tarefa de predição do risco de acidentes graves ou não-graves em trechos de rodovias brasileiras, a fim de otimizar o desempenho e a performance dos classificadores. Os dados de acidentes foram pré-processados, analisados e técnicas de seleção de atributos foram empregadas, resultando em uma base com informações sobre o dia da semana, o turno do dia em que o acidente aconteceu, o tipo da pista, o traçado da via, o sentido da rodovia, a condição meteorológica no momento do acidente e o tipo do acidente. Diferentes modelos de aprendizado de máquina foram treinados e avaliados em quatro cenários diferentes: o cenário A utiliza uma base de dados desbalanceada com o atributo “Frequência de Acidentes”, enquanto o cenário B consiste na base de dados desbalanceada sem tal atributo; o cenário C faz uso da base de dados balanceada com o atributo “Frequência de Acidentes” e o cenário D utiliza a base de dados balanceada sem este atributo. A avaliação experimental ocorreu com o emprego das métricas acurácia, precisão, revocação e medida F. Os resultados dos cenários A e B não foram relevantes ao estudo, uma vez que os classificadores não convergiram, classificando os dados em apenas uma classe: não-grave. O melhor resultado para o cenário C foi a Rede Neural MLP, que obteve 85% de acurácia, 87% de precisão, 85% de revocação e 84% de medida F. Já para o cenário D, os melhores resultados foram combinações de dois modelos diferentes: Random Forest+BernoulliNB e Logistic Regression+ExtraTreesClassifier, ambos com 84,58% de acurácia, 88,14% de precisão, 84,58% de revocação e 84,06% medida F. In order to decrease the number of road accidents, solutions to identify influencing factors of road accidents and its risk areas are being researched throughout the world. However, road accident studies depend upon its location, hence this study uses supervised machine learning techniques and automated machine learning to classify accident risk sections of brazilian federal road s in severe or not-severe, using several features. The accident data was analized, pre-processed and its features were selected using different techniques, resulting in a set of information containing the week day and time the accident happened, the road type, the road route, the road orientation, the weather condition when the accident happened and the accident type. Machine learning models were trained and evaluated in four different scenarios: scenario A used a imbalanced database with the "accident frequency" feature, while scenario B used a imbalanced database without the "accident frequency" feature; scenario C used a balanced database with the "accident frequency" feature and scenario D used a balanced database without the "accident frequency" feature. To validate the model, the accuracy, precision, recall and F-measure metrics were used. Scenarios A and B results were disregarded since all models preticted only one class: not-severe. Scenario C best result was a MLP neural network model with 85% of accuracy, 87% of precision, 85% of recall and 84% of F-measure. The best results to scenario D were two combinations of classifiers: first, the combination of Random Forest and BernoulliNB; second, the combination of Logistic Regression and ExtraTreesClassifier, both resulting in 84,58% of accuracy, 88,14% of precision, 84,58% of recall and 84,06% of F-measure.
- Published
- 2019
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.