RSL19BD at DBDC4: Ensemble of Decision Tree-Based and LSTM-Based Models

Wang, Chih-Hao; Kato, Sosuke; Sakai, Tetsuya

doi:10.1007/978-981-15-9323-9_40

“…Thus, it can be concluded that feature extraction may negatively impact on the deep learning models and deep learning models performed better than the machine learning models. In the present thesis work same results were observed showing models performed less while using decision tress classifier for features extraction [34][35][36][37]. There are several reasons for less performance with features extraction using decision tree classifier like information loss, inadequate features from decision tree classifier, overfitting or model complexity,…”

Section: Significances Bioeng Bioscisupporting

confidence: 74%

Untitled

2024

SBB

0

View full text Add to dashboard Cite

Technological development is considered one of the major parts of this recent time as it helps to improve people's quality of life and resolve their issues and challenges faced in daily life. In recent times a happy life has been considered one of the major requirements for people as most people live under stress and face several mental disorders like depression, anxiety and loneliness. In the metal disorder space, depression is a major and common disease in recent society. According to the World Health Organization (WHO), it is estimated that 5% of adults suffer from depression. Diagnosis of depression has several challenges like time consuming patient counselling, over-dependence on doctors and accuracy of diagnosis. To resolve these diagnosis issues, computer aided system solution is required with the use of machine learning tool. The objective of this research is to develop hybrid deep learning model by using CNN and LSTM. The selected dataset which was used for this study contains a dataset of 945 subjects of mental disorders and healthy control subjects. Three hybrid models were developed and compared with different sets of extracted features. Raw data was pre-processed and applied in hybrid model and at the end model validated with the unknown EEG dataset. The hybrid model with entire features of dataset reported an accuracy of 98.0% and performed superior in comparison with other two models which trained with extracted features by using decision tree classifier. The results show that the developed hybrid CNN and LSTM model is accurate, less complex and useful in detecting mental disorders including depression using EEG signals.

show abstract

“…Thus, it can be concluded that feature extraction may negatively impact on the deep learning models and deep learning models performed better than the machine learning models. In the present thesis work same results were observed showing models performed less while using decision tress classifier for features extraction [34][35][36][37]. There are several reasons for less performance with features extraction using decision tree classifier like information loss, inadequate features from decision tree classifier, overfitting or model complexity,…”

Section: Significances Bioeng Bioscisupporting

confidence: 74%

A Hybrid Neural Network for Electroencephalogram (EEG)-based Screening of Depression

Passi

2023

SBB

0

View full text Add to dashboard Cite

Technological development is considered one of the major parts of this recent time as it helps to improve people's quality of life and resolve their issues and challenges faced in daily life. In recent times a happy life has been considered one of the major requirements for people as most people live under stress and face several mental disorders like depression, anxiety and loneliness. In the metal disorder space, depression is a major and common disease in recent society. According to the World Health Organization (WHO), it is estimated that 5% of adults suffer from depression. Diagnosis of depression has several challenges like time consuming patient counselling, over-dependence on doctors and accuracy of diagnosis. To resolve these diagnosis issues, computer aided system solution is required with the use of machine learning tool. The objective of this research is to develop hybrid deep learning model by using CNN and LSTM. The selected dataset which was used for this study contains a dataset of 945 subjects of mental disorders and healthy control subjects. Three hybrid models were developed and compared with different sets of extracted features. Raw data was pre-processed and applied in hybrid model and at the end model validated with the unknown EEG dataset. The hybrid model with entire features of dataset reported an accuracy of 98.0% and performed superior in comparison with other two models which trained with extracted features by using decision tree classifier. The results show that the developed hybrid CNN and LSTM model is accurate, less complex and useful in detecting mental disorders including depression using EEG signals.

show abstract

“…Para cálculo da similaridade entre embeddings de duas sentenças, melhores resultados foram observados utilizando-se a similaridade de cosseno entre todos os pares de termos. Esta abordagem seria posteriormente aprimorada e reapresentada à competição DBDC4 com melhores resultados (Wang et al, 2019).…”

Section: Participantes Da Competição Dbdc3unclassified

Detecção de quebras em diálogos humano-computador

Andrade

¹

,

Paraboni

²

2022

Linguamática

0

View full text Add to dashboard Cite

Com o crescimento constante no uso de tecnologias de relacionamento com o consumidor na Internet, os sistemas de chatbot se tornaram onipresentes no processamento de linguagem natural (PLN) e áreas relacionadas. Apesar dos avanços significativos nos últimos anos, no entanto, sistemas desse tipo nem sempre fornecem resultados plausíveis e consistentes, em muitos casos levando a uma quebra no diálogo. Assim, há grande interesse em investigar as circunstâncias nas quais erros deste tipo são produzidos e, quando possível, aprimorar o projeto destes sistemas de modo a minimizar tais erros. Com base nestas observações, neste trabalho abordamos a questão da detecção automática de quebras em diálogos humano-computador apresentando três modelos que levam em consideração o histórico de diálogo para decidir quando ele possui maior probabilidade de culminar em uma quebra. Os modelos propostos exploram uma variedade de métodos de PLN recentes, e são avaliados tanto com base em um conjunto de dados de diálogos reais em português entre usuários humanos e sistemas de chatbot desenvolvido especificamente para este fim, como também utilizando benchmarks publicamente disponíveis para o idioma inglês.

show abstract

RSL19BD at DBDC4: Ensemble of Decision Tree-Based and LSTM-Based Models

Cited by 5 publications

References 4 publications

Untitled

Untitled

A Hybrid Neural Network for Electroencephalogram (EEG)-based Screening of Depression

Detecção de quebras em diálogos humano-computador

Contact Info

Product

Resources

About