• Phonetics and Speech Sciences
• /
• v.15 no.2
• /
• pp.53-59
• /
• 2023

Detection of children with autism spectrum disorder (ASD) based on speech has relied on predefined feature sets due to their ease of use and the capabilities of speech analysis. However, clinical impressions may not be adequately captured due to the broad range and the large number of features included. This paper demonstrates that the knowledge-driven speech features (KDSFs) specifically tailored to the speech traits of ASD are more effective and efficient for detecting speech of ASD children from that of children with typical development (TD) than a predefined feature set, extended Geneva Minimalistic Acoustic Standard Parameter Set (eGeMAPS). The KDSFs encompass various speech characteristics related to frequency, voice quality, speech rate, and spectral features, that have been identified as corresponding to certain of their distinctive attributes of them. The speech dataset used for the experiments consists of 63 ASD children and 9 TD children. To alleviate the imbalance in the number of training utterances, a data augmentation technique was applied to TD children's utterances. The support vector machine (SVM) classifier trained with the KDSFs achieved an accuracy of 91.25%, surpassing the 88.08% obtained using the predefined set. This result underscores the importance of incorporating domain knowledge in the development of speech technologies for individuals with disorders.

• Steel and Composite Structures
• /
• v.46 no.6
• /
• pp.731-744
• /
• 2023

Vehicle load information is an important role in operating and ensuring the structural health of cable-stayed bridges. In this regard, an efficient and economic method is proposed for vehicle load detection based on the observed cable tension and vehicle position using a graph neural network (GNN). Datasets are first generated using the practical advanced analysis program (PAAP), a robust program for modeling and considering both geometric and material nonlinearities of bridge structures subjected to vehicle load with low computational costs. With the superiority of GNN, the proposed model is demonstrated to precisely capture complex nonlinear correlations between the input features and vehicle load in the output. Four popular machine learning methods including artificial neural network (ANN), decision tree (DT), random forest (RF), and support vector machines (SVM) are refereed in a comparison. A case study of a cable-stayed bridge with the typical truck is considered to evaluate the model's performance. The results demonstrate that the GNN-based model provides high accuracy and efficiency in prediction with satisfactory correlation coefficients, efficient determination values, and very small errors; and is a novel approach for vehicle load detection with the input data of the existing monitoring system.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2022.06a
• /
• pp.301-301
• /
• 2022

본 연구는 운영 조건이 달라짐에 따라 회전수가 변하는 기기의 정상적 가동 여부와 고장 종류를 판별하기 위한 인공지능 알고리즘의 적용을 다루고 있다. 회전수가 변하는 장비로부터 계측된 상태 모니터링 센서의 신호는 비정상(non-stationary)적 특성이 있으므로, 상태 신호의 한계치가 고장 판별의 기준이 되기 어렵다는 점을 해결하고자 하였다. 정상 가동 여부는 이상 감지에 효율적인 오토인코더 및 기계학습 알고리즘을 적용하였으며, 고장 종류 판별에는 기계학습법과 합성곱 기반의 심층학습 방법을 적용하였다. 변하는 회전수와 연계된 주파수의 비정상적 시계열도 적절한 고장 특징 (Feature)로 대변될 수 있도록 시간 및 주파수 영역에서 특징 벡터를 구성할 수 있음을 예제로 설명하였다. 차원 축소 및 카이 제곱 기법을 적용하여 최적의 특징 벡터를 추출하여 기계학습의 분류 알고리즘이 비정상적 회전 신호를 가진 장비의 고장 예측에 활용될 수 있음을 보였다. 이 과정에서 k-NN(k-Nearest Neighbor), SVM(Support Vector Machine), Random Forest의 기계학습 알고리즘을 적용하였다. 또한 시계열 기반의 오토인코더 및 CNN (Convolution Neural Network) 적용하여 이상 감지와 고장진단을 수행한 결과를 비교하여 제시하였다.

• The Journal of Korea Robotics Society
• /
• v.19 no.2
• /
• pp.203-212
• /
• 2024

Simultaneous motion is essential in the activities of daily living (ADL). For motion intention recognition, surface electromyogram (sEMG) and corresponding motion label is necessary. However, this process is time-consuming and it may increase the burden of the user. Therefore, we propose a simultaneous motion recognition framework using data augmentation based on muscle activation model. The model consists of multiple point sources to be optimized while the number of point sources and their initial parameters are automatically determined. From the experimental results, it is shown that the framework has generated the data which are similar to the real one. This aspect is quantified with the following two metrics: structural similarity index measure (SSIM) and mean squared error (MSE). Furthermore, with k-nearest neighbor (k-NN) or support vector machine (SVM), the classification accuracy is also enhanced with the proposed framework. From these results, it can be concluded that the generalization property of the training data is enhanced and the classification accuracy is increased accordingly. We expect that this framework reduces the burden of the user from the excessive and time-consuming data acquisition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.11
• /
• pp.2784-2799
• /
• 2012

Biological data have been increased exponentially in recent years, and analyzing these data using data mining tools has become one of the major issues in the bioinformatics research community. This paper focuses on the protein construction process in higher organisms where the deoxyribonucleic acid, or DNA, sequence is filtered. In the process, "unmeaningful" DNA sub-sequences (called introns) are removed, and their meaningful counterparts (called exons) are retained. Accurate recognition of the boundaries between these two classes of sub-sequences, however, is known to be a difficult problem. Conventional approaches for recognizing these boundaries have sought for solely enhancing machine learning techniques, while inherent nature of the data themselves has been overlooked. In this paper we present an approach which makes use of the data attributes inherent to species in order to increase the accuracy of the boundary recognition. For experimentation, we have taken the data sets for four different species from the University of California Santa Cruz (UCSC) data repository, divided the data sets based on the species types, then trained a preprocessed version of the data sets on neural network(NN)-based and support vector machine(SVM)-based classifiers. As a result, we have observed that each species has its own specific features related to the splice sites, and that it implies there are related distances among species. To conclude, dividing the training data set based on species would increase the accuracy of predicting splicing junction and propose new insight to the biological research.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.64 no.1
• /
• pp.15-26
• /
• 2022

3D open-field farm model developed from UAV (Unmanned Aerial Vehicle) data could make crop monitoring easier, also could be an important dataset for various fields like remote sensing or precision agriculture. It is essential to separate crops from the non-crop area because labeling in a manual way is extremely laborious and not appropriate for continuous monitoring. We, therefore, made a 3D open-field farm model based on UAV images and developed a crop segmentation model using a supervised machine learning algorithm. We compared performances from various models using different data features like color or geographic coordinates, and two supervised learning algorithms which are SVM (Support Vector Machine) and KNN (K-Nearest Neighbors). The best approach was trained with 2-dimensional data, ExGR (Excess of Green minus Excess of Red) and z coordinate value, using KNN algorithm, whose accuracy, precision, recall, F1 score was 97.85, 96.51, 88.54, 92.35% respectively. Also, we compared our model performance with similar previous work. Our approach showed slightly better accuracy, and it detected the actual crop better than the previous approach, while it also classified actual non-crop points (e.g. weeds) as crops.

• Analytical Science and Technology
• /
• v.28 no.3
• /
• pp.187-195
• /
• 2015

The heat of sublimation (HOS) is an essential parameter used to resolve environmental problems in the transfer of organic contaminants to the atmosphere and to assess the risk of toxic chemicals. The experimental measurement of the heat of sublimation is time-consuming, expensive, and complicated. In this study, quantitative structural property relationships (QSPR) were used to develop a simple and predictive model for measuring the heat of sublimation of organic compounds. The population-based forward selection method was applied to select an informative subset of descriptors of learning algorithms, such as by using multiple linear regression (MLR) and the support vector machine (SVM) method. Each individual model and consensus model was evaluated by internal validation using the bootstrap method and y-randomization. The predictions of the performance of the external test set were improved by considering their applicability to the domain. Based on the results of the MLR model, we showed that the heat of sublimation was related to dispersion, H-bond, electrostatic forces, and the dipole-dipole interaction between inter-molecules.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.5
• /
• pp.1011-1019
• /
• 2015

CAPTCHA(Completely Automated Public Turing test to tell Computers and Humans Apart) is a test used in computing to distinguish whether or not the user is computer or human. Many web sites mostly use the character-based CAPTCHA consisting of digits and characters. Recently, with the development of OCR technology, simple character-based CAPTCHA are broken quite easily. As an alternative, many web sites add noise to make it harder for recognition. In this paper, we analyzed the most recent CAPTCHA, which incorporates the addition of the natural images to obfuscate the characters. We proposed an efficient method using support vector machine to separate the characters from the background image and use convolutional neural network to recognize each characters. As a result, 368 out of 1000 CAPTCHAs were correctly identified, it was demonstrated that the current CAPTCHA is not safe.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.1
• /
• pp.288-307
• /
• 2016

Image distortions are typically characterized by degradations of structures. Dictionaries learned from natural images can capture the underlying structures in images, which are important for image quality assessment (IQA). This paper presents a general-purpose no-reference image quality metric using a GRadient-Induced Dictionary (GRID). A dictionary is first constructed based on gradients of natural images using K-means clustering. Then image features are extracted using the dictionary based on Euclidean-norm coding and max-pooling. A distortion classification model and several distortion-specific quality regression models are trained using the support vector machine (SVM) by combining image features with distortion types and subjective scores, respectively. To evaluate the quality of a test image, the distortion classification model is used to determine the probabilities that the image belongs to different kinds of distortions, while the regression models are used to predict the corresponding distortion-specific quality scores. Finally, an overall quality score is computed as the probability-weighted distortion-specific quality scores. The proposed metric can evaluate image quality accurately and efficiently using a small dictionary. The performance of the proposed method is verified on public image quality databases. Experimental results demonstrate that the proposed metric can generate quality scores highly consistent with human perception, and it outperforms the state-of-the-arts.

• KIISE Transactions on Computing Practices
• /
• v.23 no.5
• /
• pp.322-327
• /
• 2017

The classification problem in the field of Natural Language Processing has been studied for a long time. Continuing forward with our previous research, which classifies large-scale text using Convolutional Neural Networks (CNN), we implemented Recurrent Neural Networks (RNN), Long-Short Term Memory (LSTM) and Gated Recurrent Units (GRU). The experiment's result revealed that the performance of classification algorithms was Multinomial Naïve Bayesian Classifier < Support Vector Machine (SVM) < LSTM < CNN < GRU, in order. The result can be interpreted as follows: First, the result of CNN was better than LSTM. Therefore, the text classification problem might be related more to feature extraction problem than to natural language understanding problems. Second, judging from the results the GRU showed better performance in feature extraction than LSTM. Finally, the result that the GRU was better than CNN implies that text classification algorithms should consider feature extraction and sequential information. We presented the results of fine-tuning in deep neural networks to provide some intuition regard natural language processing to future researchers.