2023
Arfilinia, A.; Andreswari, R.; Hamami, F.; Machado, J. M. F.
Multidimensional Sentiment Analysis of Tourism Object in DKI Jakarta, Banten, East Java, Central Java and West Java using Support Vector Machine Algorithm Proceedings Article
Em: Institute of Electrical and Electronics Engineers Inc., 2023, ISBN: 9798350303414, (cited By 0; Conference of 5th International Conference on Advancement in Data Science, E-Learning and Information Systems, ICADEIS 2023 ; Conference Date: 2 August 2023 Through 3 August 2023; Conference Code:193233).
Resumo | Links | BibTeX | Etiquetas: Confusion matrix; F1 scores; Labelings; Multi-dimensional sentiments; Multidimensional sentiment analyse; Sentiment analysis; Support vectors machine; Textblob; Transformer, Image resolution; Matrix algebra; Sentiment analysis, Support vector machines
@inproceedings{Arfilinia2023,
title = {Multidimensional Sentiment Analysis of Tourism Object in DKI Jakarta, Banten, East Java, Central Java and West Java using Support Vector Machine Algorithm},
author = {A. Arfilinia and R. Andreswari and F. Hamami and J. M. F. Machado},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85175628715&doi=10.1109%2fICADEIS58666.2023.10270985&partnerID=40&md5=9141e7a3e64b959a43789749f4526161},
doi = {10.1109/ICADEIS58666.2023.10270985},
isbn = {9798350303414},
year = {2023},
date = {2023-01-01},
journal = {ICADEIS 2023 - International Conference on Advancement in Data Science, E-Learning and Information Systems: Data, Intelligent Systems, and the Applications for Human Life, Proceeding},
publisher = {Institute of Electrical and Electronics Engineers Inc.},
abstract = {Information technology provides several conveniences, one of which is facilitating tourists in searching for information about tourist attractions. One of the services that can be used is Google My Business. The comments or reviews from tourists are numerous, thus requiring a significant amount of time to review them one by one. Therefore, a method is needed to address this issue, which is sentiment analysis. In this study, a multidimensional sentiment analysis was conducted on tourist attractions in the provinces of DKI Jakarta, Banten, East Java, Central Java, and West Java. Data was collected between January and March 2023 using the Data Miner tool. Two labeling techniques, namely Transformer and Textblob, were compared for labeling the data. The confusion matrix was employed as the evaluation tool, and the Support Vector Machine (SVM) technique was used to implement sentiment analysis. Labeling using transformers obtains an accuracy of 0.8169 or 81.6%, then the average precision, recall and f1-score are 68%, 54%, abd 58%. While labeling using textblob, obtained an accuracy of 0.9257 or 92.5%, the average precision is 88%, recall and f1-score results are 87%. The accuracy, precision, recall, and f1-score results indicate that labeling using Textblob outperforms the labeling using Transformers. © 2023 IEEE.},
note = {cited By 0; Conference of 5th International Conference on Advancement in Data Science, E-Learning and Information Systems, ICADEIS 2023 ; Conference Date: 2 August 2023 Through 3 August 2023; Conference Code:193233},
keywords = {Confusion matrix; F1 scores; Labelings; Multi-dimensional sentiments; Multidimensional sentiment analyse; Sentiment analysis; Support vectors machine; Textblob; Transformer, Image resolution; Matrix algebra; Sentiment analysis, Support vector machines},
pubstate = {published},
tppubtype = {inproceedings}
}
Information technology provides several conveniences, one of which is facilitating tourists in searching for information about tourist attractions. One of the services that can be used is Google My Business. The comments or reviews from tourists are numerous, thus requiring a significant amount of time to review them one by one. Therefore, a method is needed to address this issue, which is sentiment analysis. In this study, a multidimensional sentiment analysis was conducted on tourist attractions in the provinces of DKI Jakarta, Banten, East Java, Central Java, and West Java. Data was collected between January and March 2023 using the Data Miner tool. Two labeling techniques, namely Transformer and Textblob, were compared for labeling the data. The confusion matrix was employed as the evaluation tool, and the Support Vector Machine (SVM) technique was used to implement sentiment analysis. Labeling using transformers obtains an accuracy of 0.8169 or 81.6%, then the average precision, recall and f1-score are 68%, 54%, abd 58%. While labeling using textblob, obtained an accuracy of 0.9257 or 92.5%, the average precision is 88%, recall and f1-score results are 87%. The accuracy, precision, recall, and f1-score results indicate that labeling using Textblob outperforms the labeling using Transformers. © 2023 IEEE.
Neto, C.; Ferreira, D.; Cunha, H.; Pires, M.; Marques, S.; Sousa, R.; Machado, J.
Recommendation of Medical Exams to Support Clinical Diagnosis Based on Patient’s Symptoms Proceedings Article
Em: J.M., Peixoto H. Machado (Ed.): pp. 91-100, Springer Science and Business Media Deutschland GmbH, 2023, ISSN: 18678211, (cited By 0; Conference of 3rd International Conference on AI-assisted Solutions for COVID-19 and Biometrical Applications in Smart Cities, AISCOVID-19 2022 ; Conference Date: 16 November 2022 Through 18 November 2022; Conference Code:298899).
Resumo | Links | BibTeX | Etiquetas: Clinical diagnosis; CRISP-DM; Diagnoses of disease; Health care professionals; Health professionals; Machine learning techniques; Medical areas; Medical conditions; Medical errors; Medical exam, Diagnosis; Learning systems; Nearest neighbor search, Support vector machines
@inproceedings{Neto202391,
title = {Recommendation of Medical Exams to Support Clinical Diagnosis Based on Patient’s Symptoms},
author = {C. Neto and D. Ferreira and H. Cunha and M. Pires and S. Marques and R. Sousa and J. Machado},
editor = {Peixoto H. Machado J.M.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85172725999&doi=10.1007%2f978-3-031-38204-8_8&partnerID=40&md5=e29bde9aad6cf908bce88ef83007236a},
doi = {10.1007/978-3-031-38204-8_8},
issn = {18678211},
year = {2023},
date = {2023-01-01},
journal = {Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST},
volume = {485 LNICST},
pages = {91-100},
publisher = {Springer Science and Business Media Deutschland GmbH},
abstract = {Nowadays, it is essential that the error in the decisions made by health professionals is as small as possible. This applies to any medical area, including the recommendation of medical exams based on certain symptoms for the diagnosis of diseases. This study aims to explore the use of different Machine Learning techniques to increase the confidence of the medical exams prescribed by healthcare professionals. A successful implementation of this proposal could reduce the probability of medical errors in what concerns the prescription of medical exams and, consequently, the diagnosis of medical conditions. Thus, in this paper, six Machine Learning models were applied and optimized, namely, RF, DT, k-NN, NB, SVM and RNN, in order to find the most suitable model for the problem at hand. The results obtained with this study were promising, achieving high accuracy values with RF, DT and k-NN. © 2023, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.},
note = {cited By 0; Conference of 3rd International Conference on AI-assisted Solutions for COVID-19 and Biometrical Applications in Smart Cities, AISCOVID-19 2022 ; Conference Date: 16 November 2022 Through 18 November 2022; Conference Code:298899},
keywords = {Clinical diagnosis; CRISP-DM; Diagnoses of disease; Health care professionals; Health professionals; Machine learning techniques; Medical areas; Medical conditions; Medical errors; Medical exam, Diagnosis; Learning systems; Nearest neighbor search, Support vector machines},
pubstate = {published},
tppubtype = {inproceedings}
}
Nowadays, it is essential that the error in the decisions made by health professionals is as small as possible. This applies to any medical area, including the recommendation of medical exams based on certain symptoms for the diagnosis of diseases. This study aims to explore the use of different Machine Learning techniques to increase the confidence of the medical exams prescribed by healthcare professionals. A successful implementation of this proposal could reduce the probability of medical errors in what concerns the prescription of medical exams and, consequently, the diagnosis of medical conditions. Thus, in this paper, six Machine Learning models were applied and optimized, namely, RF, DT, k-NN, NB, SVM and RNN, in order to find the most suitable model for the problem at hand. The results obtained with this study were promising, achieving high accuracy values with RF, DT and k-NN. © 2023, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.