• Phonetics and Speech Sciences
• /
• v.15 no.4
• /
• pp.71-80
• /
• 2023

This study primarily aimed to develop an automated stuttering identification and classification method using artificial intelligence technology. In particular, this study aimed to develop a deep learning-based identification model utilizing the convolutional neural networks (CNNs) algorithm for Korean speakers who stutter. To this aim, speech data were collected from 9 adults who stutter and 9 normally-fluent speakers. The data were automatically segmented at the phrasal level using Google Cloud speech-to-text (STT), and labels such as 'fluent', 'blockage', prolongation', and 'repetition' were assigned to them. Mel frequency cepstral coefficients (MFCCs) and the CNN-based classifier were also used for detecting and classifying each type of the stuttered disfluency. However, in the case of prolongation, five results were found and, therefore, excluded from the classifier model. Results showed that the accuracy of the CNN classifier was 0.96, and the F1-score for classification performance was as follows: 'fluent' 1.00, 'blockage' 0.67, and 'repetition' 0.74. Although the effectiveness of the automatic classification identifier was validated using CNNs to detect the stuttered disfluencies, the performance was found to be inadequate especially for the blockage and prolongation types. Consequently, the establishment of a big speech database for collecting data based on the types of stuttered disfluencies was identified as a necessary foundation for improving classification performance.

• Journal of The Korean Association For Science Education
• /
• v.44 no.2
• /
• pp.141-153
• /
• 2024

The 2022 revised science curriculum aims to develop the ability to solve scientific problems arising in daily life and society based on convergent thinking stimulated through participation in research activities using artificial intelligence (AI). Therefore, we developed a science-AI convergence education program that combines the science curriculum with artificial intelligence and employed it in convergence classes for high school students. The aim of the science-AI convergence class was for students to qualitatively understand the movement of a damped pendulum and build an AI model to predict the position of the pendulum using the block coding platform KNIME. Individual in-depth interviews were conducted to understand and interpret the learners' experiences. Based on Giorgi's phenomenological research methodology, we described the learners' learning processes and changes, challenges and limitations of the class. The students collected data and built the AI model. They expected to be able to predict the surrounding phenomena based on their experimental results and perceived the convergence class positively. On the other hand, they still perceived an with the unfamiliarity of platform, difficulty in understanding the principle of AI, and limitations of the teaching method that they had to follow, as well as limitations of the course content. Based on this, we discussed the strengths and limitations of the science-AI convergence class and made suggestions for science-AI convergence education. This study is expected to provide implications for developing science-AI convergence curricula and implementing them in the field.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.2
• /
• pp.109-122
• /
• 2003

As XML documents become popular in the World Wide Web, a lot of research is being done on XML storage systems that store and manage XML documents using existing DBMSS. However, most of them have been done in the context of relational DBMSS rather than object-relational DBMSS, which have more powerful modeling capability than relational ones. In this paper, we design and implement an XML storage system, ODYSSEUS/XMLStore, for the ODYSSEUS object-relational DBMS that has been under development at KAIST. First, we analyze the mapping from the structure of XML documents to the relational or object-relational database schema. Second, we propose a method for describing the mapping analyzed using a standard language, XML Schema. Third, we propose a storage structure for storing the mapping information specified by the users in the database. Fourth, we propose detailed algorithms for storing XML documents in the relational or object-relational database based on the mapping information provided by the users.

• Journal of KIISE
• /
• v.44 no.3
• /
• pp.332-337
• /
• 2017

When using IFC data consisting of STEP schema based on the EXPRESS language, it is not easy for collaborating project stakeholders to share BIM modeling shape information. The IFC viewer application must be installed on the desktop PC to review the BIM modeling shape information defined within the IFC, because the IFC viewer application not only parse STEP structure information model but also process the 3D feature construction for a 3D visualization. Therefore, we propose a lightweight data structure design for web visualization by parsing IFC data and constructing 3D modeling data. Our experimental results show the weight reduction of IFC data is about 40% of original file size and the web visualization is able to see the same quality with all web browsers which support WebGL on PCs and smartphones. If applied research is conducted about the web visualization based on IFC data of the last construction phase, it could be utilized in various fields ranging from the facility maintenance to indoor location-based services.

• KIISE Transactions on Computing Practices
• /
• v.23 no.1
• /
• pp.28-36
• /
• 2017

After the genome projects of the 90s, a vast number of gene studies have been stored in online databases. By using these databases, several biological relationships can be inferred. In this study, we proposed a method to infer disease-gene relationships using title and body in biomedical text. The title was used to extract hub genes from data in the literature; whereas, the body of the literature was used to extract sub genes that are related to hub genes. Through these steps, we were able to construct a local gene-network for each report in the literature. By integrating the local gene-networks, we then constructed a global gene-network. Subsequent analyses of the global gene-network allowed inference of disease-related genes with high rank. We validated the proposed method by comparing with previous methods. The results indicated that the proposed method is a meaningful approach to infer disease-related genes.

• Current Research on Agriculture and Life Sciences
• /
• v.32 no.2
• /
• pp.79-84
• /
• 2014

The precision aerial application of agricultural unmanned helicopters has become a new paradigm for small farms with orchards, paddy, and upland fields. The needs of agricultural applications require easy and affordable control systems. Recent developments of MEMS technology based on inertial sensors and high speed DSP have enabled the fabrication of low-cost attitude system. Therefore, this study evaluates inertial MEMS sensors for estimating the attitude of an agricultural unmanned helicopter. The accuracies and errors of gyro and acceleration sensors were verified using a pendulum system. The true motion values were calculated using a theoretical estimation and absolute encoder measurement of the pendulum, and then the sensor output was compared with reference values. When comparing the sensor measurements and true values, the errors were determined to be 4.32~5.72%, 3.53~6.74%, and 3.91~4.16% for the gyro rate and x-, z- accelerations, respectively. Thus, the measurement results confirmed that the inertial sensors are effective for establishing an attitude and heading reference system (AHRES). The sensors would be constructed in gimbals for the estimating and proving attitude measurements in the following paper.

• Science of Emotion and Sensibility
• /
• v.20 no.1
• /
• pp.67-74
• /
• 2017

The goal of this study is to present quantification method for empathy. The micro-movement technology (non-contact sensing method) was used to identify empathy level. Participants were first divided into two groups: Empathized and not empathized. Then, the upper body data of participants were collected utilizing web-cam when participants carried expression tasks. The data were analyzed and categorized into 0.5 Hz, 1 Hz, 3 Hz, 5 Hz, 15 Hz. The average movement, variation, and synchronization of the movement were then compared. The results showed a low average movement and variation in a group who empathized. Also, the participants, who empathized, synchronized their movement during the task. This indicates that the people concentrates with each other when empathy has been established and show different levels of movement. These findings suggest the possibility of empathy quantification using non-contact sensing method.

• Journal of the Korean Society for Library and Information Science
• /
• v.44 no.3
• /
• pp.81-102
• /
• 2010

The purpose of this study was to investigate college students' preferences of Web-based OPAC retrieval techniques and their ABO blood types as an empirical survey. Data was collected through a self-designed questionnaire with a total of 101 undergraduate students from the College of Social Sciences responding. The collected data was analyzed using descriptive statistics, and One-way ANOVA. The results show that 'title' was most preferred among the access points, 'AND' was the most preferred Boolean operator, 'publication year' and 'subject' were the most favored techniques in limiting the scope of retrieval, and 'record number limit per page' was the most frequently used for displaying retrieval results. The results also show that there were little(3 out of 22, i.e. 13.6%) statistically significant differences between the college students' preferences of Web-based OPAC techniques and their blood type.

• Journal of the HCI Society of Korea
• /
• v.8 no.1
• /
• pp.19-28
• /
• 2013

As yet the use of 3D scanning technology has been limited to documentation, preservation and monitoring in cultural and natural heritage domain. Appropriate visualization of precise geometric information in scan data can support scientific interpretation of the domain experts. We studied the rendering techniques which visualize the required features from scanned models, and proposed a 3D scan data visualization pipeline, rendering methods, and a classification scheme. As a case study, we analyzed the traditional method in the study of dinosaur tracks and performed the visualization of 3D scanned models. A user test based on the result confirmed an effectiveness of the proposed method. This research showed a practical method in which 3D scanned models can be used to effective interpretation of heritage.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.81-88
• /
• 2021

Convertible bonds are financial products that contain the nature of both bonds and shares, which are generally issued by companies with lower credit ratings to increase liquidity. Conversion bonds rely on qualitative judgment in the past, although decision-making on whether and when to exercise the right to convert is the most important issue. Therefore, this paper proposes to apply artificial neural network techniques to scientifically determine the exercise of conversion rights. We distinguish between a total of 1,800 learning data published in the past and 200 predictive experimental data and build an artificial neural network learning model. As a result, the parity performance in most groups was excellent, achieving an average excess of about 10% or more. In particular, groups 3-6 recorded an average excess of about 20% and group 6 recorded an average excess of about 37%. This paper is meaningful in that it focused on solving decision problems by converging and applying machine learning techniques, a representative technology of the fourth industry, to the financial sector.