論文題目集成和多模態學習用於病理性語音分類;ENSEMBLE AND MULTIMODAL LEARNING FOR PATHOLOGICAL VOICE CLASSIFICATION

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/84131

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/84131

Title:	論文題目集成和多模態學習用於病理性語音分類;ENSEMBLE AND MULTIMODAL LEARNING FOR PATHOLOGICAL VOICE CLASSIFICATION
Authors:	黎亞媞;Ariyanti, Whenty
Contributors:	資訊工程學系
Keywords:	病理性語音;聲學信號;集成學習;二進制分類;Pathological Voice;Acoustic Signal;Ensemble Learning;Binary Classification
Date:	2020-08-20
Issue Date:	2020-09-02 18:09:42 (UTC+8)
Publisher:	國立中央大學
Abstract:	語音障礙是現代社會中最常見的醫學疾病之一，特別是對於有職業語音需求的人群。在本文中，我們研究了一種通過組合聲信號和病歷對病理性語音障礙進行分類的堆疊式集成學習方法。在提出的集成學習框架中，堆疊支持向量機（SVM）形成了一組弱分類器，並為元學習者提供了一個深度神經網絡（DNN）。基於DNN的高度複雜性，將聲學特徵和病歷結合起來以獲得更好的分類性能。與單個SVM和DNN分類器相比，具有更好的性能，並且具有顯著的優勢。;Voice disorders are one of the most common medical diseases in modern society, especially for those with occupational voice demand. In this paper, we investigate a stacked ensemble learning method to classify pathological voice disorder by combining acoustic signals and medical records. In the proposed ensemble learning framework, a stacked support vector machine (SVM) form a set of weak classifiers and a deep neural network (DNN) for a meta learner. Based on the high complexity of DNN, acoustic features and medical records are combined to attain better classification performance. The better performance than single SVM and DNN classifiers with a notable margin.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	194	View/Open

社群 sharing

Loading...