• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.365-370
• /
• 2023

This paper aims to compare the performance of speech data classification into two groups, adult and elderly, based on the acoustic and linguistic characteristics that change due to aging, such as changes in respiratory patterns, phonation, pitch, frequency, and language expression ability. For acoustic features we used attributes related to the frequency, amplitude, and spectrum of speech voices. As for linguistic features, we extracted hidden state vector representations containing contextual information from the transcription of speech utterances using KoBERT, a Korean pre-trained language model that has shown excellent performance in natural language processing tasks. The classification performance of each model trained based on acoustic and linguistic features was evaluated, and the F1 scores of each model for the two classes, adult and elderly, were examined after address the class imbalance problem by down-sampling. The experimental results showed that using linguistic features provided better performance for classifying adult and elderly than using acoustic features, and even when the class proportions were equal, the classification performance for adult was higher than that for elderly.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.2A
• /
• pp.161-168
• /
• 2010

The bonded tendon was adopted to the reactor building of some operating nuclear power plants in Korea and the assessment of the effective prestress force on the bonded tendon is being issued as an important pending problem for continuous operation beyond their design life. The sensitivity analysis of various parameters was carried out to evaluate the effective prestress force using the system identification technique and the optimal parameters were determined for SI technique in this study. The 1/5 scaled post-tensioned concrete beams with the bonded tendon type were manufactured and in order to investigate the relationship of the natural frequency and the displacement to the effective prestress force, impact test, SIMO sine sweep test and bending test using the optical fiber sensor and the compact displacement transducer were carried out. As a result of tests, both the natural frequency and the displacement show the good relationship with the effective prestress force and both parameters are available for the SI technique to estimate the effective prestress force.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.2
• /
• pp.78-87
• /
• 2006

In this paper. we propose an effective mask estimation scheme for missing-feature reconstruction in order to achieve robust speech recognition under unknown noise environments. In the previous work. colored noise is used for training the mask classifer, which is generated from the entire frequency Partitioned signals. However it gives a limited performance under the restricted number of training database. To reflect the spectral events of more various background noise and improve the performance simultaneously. a new Bayesian classifier for mask estimation is proposed, which works independent of other frequency bands. In the proposed method, we employ the colored noise which is obtained by combining colored noises generated from each frequency band in order to reflect more various noise environments and mitigate the 'sparse' database problem. Combined with the cluster-based missing-feature reconstruction. the performance of the proposed method is evaluated on a task of noisy speech recognition. The results show that the proposed method has improved performance compared to the Previous method under white noise. car noise and background music conditions.

• Smart Structures and Systems
• /
• v.33 no.2
• /
• pp.145-163
• /
• 2024

The evaluation of dynamic characteristics of bridges under operational traffic loads is a crucial aspect of bridge structural health monitoring. In the vehicle-bridge interaction (VBI) system, the vibration responses of bridge exhibit time-varying characteristics. To address this issue, an accurate time-frequency analysis method that combines the autoregressive power spectrum based empirical wavelet transform (AR-EWT) and local maximum synchrosqueezing transform (LMSST) is proposed to identify the time-varying instantaneous frequencies (IFs) of the bridge in the VBI system. The AR-EWT method decomposes the vibration response of the bridge into mono-component signals. Then, LMSST is employed to identify the IFs of each mono-component signal. The AR-EWT combined with the LMSST method (AR-EWT+LMSST) can resolve the problem that LMSST cannot effectively identify the multi-component signals with weak amplitude components. The proposed AR-EWT+LMSST method is compared with some advanced time-frequency analysis techniques such as synchrosqueezing transform (SST), synchroextracting transform (SET), and LMSST. The results demonstrate that the proposed AR-EWT+LMSST method can improve the accuracy of identified IFs. The effectiveness and applicability of the proposed method are validated through a multi-component signal, a VBI numerical model with a four-degree-of-freedom half-car, and a VBI model experiment. The effect of vehicle characteristics, vehicle speed, and road surface roughness on the identified IFs of bridge are investigated.

• Journal of agricultural medicine and community health
• /
• v.29 no.1
• /
• pp.147-161
• /
• 2004

Objectives: There are many habitual drinking in rural area. So it is the key point of drinking control policy in rural community to understand the drinking behavior in leisure time and to have an appropriate screening method for problem drinking. CAGE and AUDIT are famous screening tools for problem drinking and alcoholics. Even though there are some studies to validate CAGE and AUDIT which translated in Korean, they were not studied with community based population but with hospital based patients. In this study we assessed the usefulness of CAGE and AUDIT for selecting problem drinking in a rural population, and compared problem drinkers with normal group about spending leisure, Methods: The study subjects were 120 residents over 20 years old who lived in 3 districts in Dong-San Myun near Chun-chon city. We made up questionnaire by interview from Feb. 13 to 19, 2004. Results: The mean age of study population was 66.01 .26 years old. Defining the problem drinking as more than 12 score in AUDIT and more than 2 score in CAGE, the proportion of problem drinker was 30.600 and 28.9% respectively. This proportions were higher than those of other national wide studies. There were significant difference in drinking frequency per week and amount per one episode between problem drinker and normal group. Experience about driving, accident, injury, disturbance in working and quarrel after drinking were also significantly different. Problem drinker were more tolerable to the bad social culture about drinking (eg. force to drink, bad drunken habit. overdrinking, drinking relay etc.) than normal group. Watching TV and playing with neighborhood were most frequent method of spending leisure in this study population, normal male group exercised more frequently in leisure time than problem drinker. Conclusions: It may be useful to use CAGE and AUDIT score for screening problem drinking in rural community. Appropriate utilization of leisure time may he important for control of problem drinking in rural area.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.36 no.7
• /
• pp.840-848
• /
• 2007

This study was performed in order to investigate the stress levels, sleep, anthropometric measurement and nutrient intakes their correlations of college students. General characteristics, stress scores using a stress test, sleep scores using a sleep test, anthropometric measurement, body composition using Inbody 3.0 and nutrient intakes using 3-day recalls were measured in 353 subjects (183 males and 170 females). Mean total stress score of the male students was 68.2${\pm}$45.0 and that of the female students was 86.5${\pm}$48.7. Stress for study, future, economic and value-related factors were higher than other factors in both male and female students. Female students were more stressed than male in friend, family, study, future and value-related factors. Mean sleeping time and sleep scores of the male students were 7.0 hrs and 40.6${\pm}$5.7, respectively and that of the female students were 7.1 hrs and 41.5${\pm}$5.3, respectively. Mean height, weight, percent body fat (PBF) and waist-hip ratio (WHR) of the male students were 174.8 cm, 73.3 kg, 19.0% and 0.84, respectively. Mean height, weight, PBF, and WHR of the female students were 161.7 cm, 55.3 kg, 28.7% and 0.81, respectively. Mean energy and protein intakes of the male students were 2026.3 kcal (77.9% EER) and 83.0 g (150.9% RI) and those of the female students were 1538.2 kcal (73.2% EER) and 60.7 g (134.9% RI), respectively. In male students, sleep duration, professor and future problem showed significantly negative correlation (p<0.05). For both male and female students, in correlations between sleep scores, life stress experience frequency and importance, the total scores showed significantly negative correlation (p<0.01). Sleep scores (sleep quality) have more significant correlation than sleep duration in life stress. In male students, correlations between economic problem and weight, waist circumference and hip circumference showed significantly negative correlation. In female students, correlations between different gender problem and body mass index (p<0.05), PBF (p<0.01), WHR (p<0.01) and obesity degree (p<0.05) showed significantly negative correlation while correlation between study problem and PBF (p<0.05), WHR (p<0.05) showed significantly positive correlation. In male students, there were significantly positive correlations between life stress experience frequency and carbohydrate calcium, iron, vitamin A; correlation between importance and calcium, iron as well as correlation between total life stress scores and iron (p<0.05). In female students, correlation between life stress experience frequency and thiamin along with correlation between importance and thiamin showed significant negative correlation (p<0.05). College students need to practice good life habits for the purpose of correctly managing life stress.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.10
• /
• pp.4619-4628
• /
• 2012

The purpose of this study was to investigate the relationship between alcohol use and drinking problems among college students. Data were collected by structured questionnaires from 400 college students by convenient sampling methods from September 26 to October 14, 2011. Data were analyzed by SPSS/WIN 15.0 program. The results were as follows: According to the Alcohol Use Disorder Identification Test (AUDIT), subjects were classified into four groups such as 40.2% of no problem, 36.7% of drinking problem, 6.8% of high risk drinking problem and 16.3% of alcohol dependence. The alcohol use and drinking problems were significant difference for gender, age, type of residence, age of first drinking, cause of drinking, frequency and amount of drinking. The alcohol use was correlated with drinking problems: social function impairment(r=.601, p<.01), behavior problem(r=.579, p<.01), family and interpersonal impairment(r=.637, p<.01). The influencing factors of drinking problems were alcohol use(${\beta}$=.640, p=.000). This study concludes that it is necessary to establish alcohol use policies on campus and develop temperance education program to reduce alcohol consumption and drinking problems.

• Journal of the Korean Dietetic Association
• /
• v.18 no.2
• /
• pp.97-114
• /
• 2012

This study was performed to assess the nutritional status and hair mineral content of children with behavioral problems and compare the values with a gender, age-matched control group. The subjects were recruited from S elementary school children in Seoul, Korea. Students scored higher than 60 points were diagnosed with behavioral problems according to the Korean-Child Behavior Checklist. Nutritional assessment of the two groups (behavior problem group: male n=15, female n=24; $9.6{\pm}1.6$ years, control group: male n=16, female n=18; $9.5{\pm}1.9$ years) was performed using a nutritional survey and by measuring hair mineral contents. In the results, food frequency questionnaire analysis showed that the intakes of anchovy (P<0.05), soybean curb (P<0.01), radish (P<0.05), bean sprouts (P<0.05), spinach (P<0.05), carrot (P<0.05), pumpkin (P<0.05), lettuce (P<0.05), cabbage (P<0.01), apple (P<0.05), and milk (P<0.01) were higher in the control group than the behavior problem group, whereas intakes of ramyeon (P<0.05), cookies (P<0.05), and coke (P<0.01) were higher in the behavior problem group than the control group. Intakes of most nutrients such as plant-derived protein (P<0.05), fiber (P<0.05), plant-derived calcium (P<0.05), phosphorus (P<0.05), plant-derived iron (P<0.05), vitamin B2 (P<0.05), vitamin B6 (P<0.05), vitamin C (P<0.01), vitamin E (P<0.05), and folate (P<0.05) were significantly higher in the control group than the behavior problem group. Hair analysis showed that the levels of arsenic (P<0.05), mercury (P<0.001), uranium (P<0.05), iron (P<0.001), boron (P<0.01), and germanium (P<0.001) were lower, but the levels of phosphate (P<0.05), chromium (P<0.001), sodium (P<0.05), and sulfur (P<0.001) were higher in the behavior problem group than the control group. Conclusively, behavioral problems constitute a complicated condition in which nutritional factors may play major roles. However, it is still under investigation as to whether or not modification of dietary habits or nutritional supplementation can improve children's behavior, since symptoms require a broad understanding of the environmental and genetic interactions.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2005.09a
• /
• pp.267-271
• /
• 2005

The goal of data mining is to extract new and useful knowledge from large scale datasets. As the amount of available data grows explosively, it became vitally important to develop faster data mining algorithms for various types of data. Recently, an interest in developing data mining algorithms that operate on graphs has been increased. Especially, mining frequent patterns from structured data such as graphs has been concerned by many research groups. A graph is a highly adaptable representation scheme that used in many domains including chemistry, bioinformatics and physics. For example, the chemical structure of a given substance can be modelled by an undirected labelled graph in which each node corresponds to an atom and each edge corresponds to a chemical bond between atoms. Internet can also be modelled as a directed graph in which each node corresponds to an web site and each edge corresponds to a hypertext link between web sites. Notably in bioinformatics area, various kinds of newly discovered data such as gene regulation networks or protein interaction networks could be modelled as graphs. There have been a number of attempts to find useful knowledge from these graph structured data. One of the most powerful analysis tool for graph structured data is frequent subgraph analysis. Recurring patterns in graph data can provide incomparable insights into that graph data. However, to find recurring subgraphs is extremely expensive in computational side. At the core of the problem, there are two computationally challenging problems. 1) Subgraph isomorphism and 2) Enumeration of subgraphs. Problems related to the former are subgraph isomorphism problem (Is graph A contains graph B?) and graph isomorphism problem(Are two graphs A and B the same or not?). Even these simplified versions of the subgraph mining problem are known to be NP-complete or Polymorphism-complete and no polynomial time algorithm has been existed so far. The later is also a difficult problem. We should generate all of 2$^n$ subgraphs if there is no constraint where n is the number of vertices of the input graph. In order to find frequent subgraphs from larger graph database, it is essential to give appropriate constraint to the subgraphs to find. Most of the current approaches are focus on the frequencies of a subgraph: the higher the frequency of a graph is, the more attentions should be given to that graph. Recently, several algorithms which use level by level approaches to find frequent subgraphs have been developed. Some of the recently emerging applications suggest that other constraints such as connectivity also could be useful in mining subgraphs : more strongly connected parts of a graph are more informative. If we restrict the set of subgraphs to mine to more strongly connected parts, its computational complexity could be decreased significantly. In this paper, we present an efficient algorithm to mine frequent subgraphs that are more strongly connected. Experimental study shows that the algorithm is scaling to larger graphs which have more than ten thousand vertices.

• Journal of The Korean Association For Science Education
• /
• v.25 no.4
• /
• pp.526-532
• /
• 2005

The purpose of this study was to analyze physics problem solving processes to qualitative and quantitative problems in the area of 'Force and Motion' in high school science. The students who have already learned the area of 'Force and Motion' during the first semester of 10th grade have taken physics test to choose students who have basic knowledge of physics. Eight students were selected. After explaining the purpose and the procedure of this study, think-aloud method was instructed to the students, and the students practiced it. After that, the students solved three problems in each quantitative and qualitative type. Then, the questionnaire of belief system on physics and physics problem solving and the prerequisite knowledge test were administered. By recording the students' solving processes, protocol was made and analyzed. After solving problems, the students expressed their confidence, intimacy, and preference. Quantitative problems needed much time at planning step than qualitative problems did. Moreover, solving time was longer and repeating frequency was more than those of qualitative problems. It seemed because even though the students qualitatively knew the answer, they should determine the given quantitative conditions, consider formulae, and recall the specific numbers. Since the students usually got access to many quantitative items in their physics study, they were accustomed to solve problems by using formulae. In addition, they put confidence in formulae, so they tended to solve problems quantitatively. As the result, they preferred quantitative problems to qualitative problems.