This report summarize the duties and result conducted in the work placement. The project in the work placement was about applying machine learning to mobile applications, which analyzed human language to forecast speakers’ level of dementia. It can be split into four parts, including collecting audio data, exploratory data analysis, machine learning analysis, and applying the machine learning model to the existing product. Firstly, collected dementia group and normal group audio data samples as the input of further analysis usage. In the second part, explored the data to find insights and filter noise before building machine learning models. After that building machine learning models analyzed the characteristics of the speech and forecast speakers' level of dementia. At last, combining the machine learning results in the mobile application development .

This report summarize the duties and result conducted in the work placement. The project in the work placement was about applying machine learning to mobile applications, which analyzed human language to forecast speakers’ level of dementia. It can be split into four parts, including collecting audio data, exploratory data analysis, machine learning analysis, and applying the machine learning model to the existing product. Firstly, collected dementia group and normal group audio data samples as the input of further analysis usage. In the second part, explored the data to find insights and filter noise before building machine learning models. After that building machine learning models analyzed the characteristics of the speech and forecast speakers' level of dementia. At last, combining the machine learning results in the mobile application development .

Applying Machine Learning Techniques to forecast the level of dementia from spontaneous speech conversations

CHEUNG, HAU YEE
2021/2022

Abstract

This report summarize the duties and result conducted in the work placement. The project in the work placement was about applying machine learning to mobile applications, which analyzed human language to forecast speakers’ level of dementia. It can be split into four parts, including collecting audio data, exploratory data analysis, machine learning analysis, and applying the machine learning model to the existing product. Firstly, collected dementia group and normal group audio data samples as the input of further analysis usage. In the second part, explored the data to find insights and filter noise before building machine learning models. After that building machine learning models analyzed the characteristics of the speech and forecast speakers' level of dementia. At last, combining the machine learning results in the mobile application development .
2021
Applying Machine Learning Techniques to forecast the level of dementia from spontaneous speech conversations
This report summarize the duties and result conducted in the work placement. The project in the work placement was about applying machine learning to mobile applications, which analyzed human language to forecast speakers’ level of dementia. It can be split into four parts, including collecting audio data, exploratory data analysis, machine learning analysis, and applying the machine learning model to the existing product. Firstly, collected dementia group and normal group audio data samples as the input of further analysis usage. In the second part, explored the data to find insights and filter noise before building machine learning models. After that building machine learning models analyzed the characteristics of the speech and forecast speakers' level of dementia. At last, combining the machine learning results in the mobile application development .
machine learning
dementia
spontaneous speech
File in questo prodotto:
File Dimensione Formato  
thesis.pdf

accesso aperto

Dimensione 2.31 MB
Formato Adobe PDF
2.31 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/34895