• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.755-759
• /
• 2009

The ability to recognize human emotion is one of the hallmarks of human-robot interaction. Especially, speaker-independent emotion recognition is a challenging issue for commercial use of speech emotion recognition systems. In general, speaker-independent systems show a lower accuracy rate compared with speaker-dependent systems, as emotional feature values depend on the speaker and his/her gender. Hence, this paper describes the realization of speaker-independent emotion recognition by rejection using confidence measure to make the emotion recognition system be homogeneous and accurate. From comparison of the proposed methods with conventional method, the improvement and effectiveness of proposed methods were clearly confirmed.

• The Korean Journal of Applied Statistics
• /
• v.16 no.2
• /
• pp.377-393
• /
• 2003

Several confidence interval estimates for the difference of two binomial proportions were introduced. Bootstrap confidence interval is also suggested. We examined the over estimation property of approximate intervals and under estimation trend of exact intervals for the difference of proportions. We compared these confidence intervals based on the average coverage probability, expected width and skewness measure. Particularly actual coverage probability were calculated by using the prior distribution of parameters. Monte Carlo simulation for small sample size is conducted. Some interesting contour plots of average coverage probability and marginal plots for several interval estimates are presented.

• International Journal of Advanced Culture Technology
• /
• v.9 no.4
• /
• pp.162-168
• /
• 2021

This is a group pretest-posttest design study that aim to evaluate the effect of problem-based learning using flipped learning to self leadership, nursing leadership, and confidence of nursing students. The subjects were 87 4th nursing students who took nursing management from March 15 to April 26, 2021. A questionnaire was used to measure self leadership, nursing leadership, and confidence. The measurement of self leadership is a 5-point Likert scale that employs a questionnaire. Flipped learning uploaded a 5-8 minute lecture video, and the subjects learned on learning management system, using smart device. PBL teaching was developed about personal growth, cooperation, nursing excellence, creative problem solving, influence. As a result of the study, self leadership was significantly improved after education than before education(t=-6.27, p<.001). Nursing leadership and confidence were significantly changed before and after education(t=-16.10, p<.001; t=-2.37, p<.001). According to the results of this study, the PBL teaching method using flipped learning is an effective teaching and learning strategy for improving leadership, and confidence. In future studies, it will be necessary to develope and verify specific teaching and learning methods by applying nursing experts in consideration of the type of leadership.

• Speech Sciences
• /
• v.9 no.1
• /
• pp.49-59
• /
• 2002

This paper concerns an effective text-dependent speaker verification method to increase the performance of speaker verification. While various speaker verification methods have already been developed, their effectiveness has not yet been formally proven in terms of achieving acceptable performance levels. This paper proposes a weighted likelihood procedure along with a confidence measure based on subword-based text-dependent speaker verification. Our aim is to remedy the low performance problem in speaker verification by exploring a means to strengthen the verification likelihood via subword-based hypothesis criteria and weighted likelihood method. Experimental results show that the proposed speaker verification method outperforms that of the speaker verification scheme without using the proposed decision by a factor of up to 1.6 times. From these results, the proposed speaker verification method is shown to be very effective and to achieve a reliable performance.

• MALSORI
• /
• no.46
• /
• pp.103-115
• /
• 2003

In this paper, we proposed Background Model Set algorithm used in the speaker verification to improve calculating confidence measure(CM) in speech recognition. CM is to display relative likelihood between recognized models and antiphone models. In previous method calculating of CM, we calculated probability and standard deviation using all phonemes in composition of antiphone models. At this process, antiphone CM brought bad recognition result. Also, recognition time increases. In order to solve this problem, we studied about method to reconstitute average and standard deviation using BMS algorithm in CM calculation.

• Journal of The Korean Astronomical Society
• /
• v.21 no.2
• /
• pp.133-153
• /
• 1988

Linear polarization has been measured for fifteen extragalactic radio sources in the field of supernova remnant S 147. These observations were made at 4885, 4835, 1665, 1515, 1465, and 1385 MHz using the Very Large Array, primarily to determine unambiguous Rotation Measures (RM) of the sources. This yields a total of 11 new RM sources. Comparisons of a sample of sources which are seen through S 147 were made with sources located farther away. The result tentatively indicates that the distribution of the rotation measure of the former population is broadened, with more than a 85% level of confidence. This constitutes evidence that there is a SNR contribution to rotation measure in sources seen through the SNR. Limits on this RM contribution are statistically at 30 < | RM | < 70 rad $m^{-2}$ with a 85% statistical level of confidence. These result imply the magnetic field in the rim of S 147 has strength 10 microguass if the electron density is $1\;cm^{-3}$.

• Speech Sciences
• /
• v.15 no.4
• /
• pp.85-96
• /
• 2008

Automatic speech recognition has severe performance degradation under noisy environments. To cope with the noise problem, many methods have been proposed. Most of them focused on noise-robust features or model adaptation. However, researchers have overlooked utterance verification (UV) under noisy environments. In this paper we discuss UV problems based on the normalized confidence measure. First, we show that UV performance is also degraded in noisy environments with the experiments of an isolated word recognition. Then we observe how the degradation of UV performances is suffered. Based on the UV experiments we propose a modeling method of the statistics of phone confidences using sigmoid functions. For obtaining the parameters of the sigmoidal models, the particle swarm optimization (PSO) is adopted. The proposed method improves 20% rejection performance. Our experimental results show that the PSO-NCM can apply noise speech recognition successfully.

• Journal of Korean Society for Quality Management
• /
• v.18 no.1
• /
• pp.40-47
• /
• 1990

The two-sample problem where an experimental treatment is compared with a control is considered. Without making any parametric-model assumptions for the distributions(or survival distributions), a measure for summarizing the differences between the treatment and the control is introduced. A method for constructing a confidence interval for the proposed measure is given in cases of complete and right random censored data. This method is illustrated with two numerical examples.

• Proceedings of the KSPS conference
• /
• 2003.05a
• /
• pp.79-82
• /
• 2003

In this paper, we proposed Backgorund Model Set algorithm for the speaker verification to improve the shortcoming of calculating process in conventional confidence measure(CM). CM is to display relative likelihood between recognized models and unrecognized models. Unrecognized models is known as antiphone models. Calculate probability and standard deviation using all phonemes at process that compose antiphone model. At this process, antiphone CM brought bad result. Also, recognition time increases. In order problem, we studied about method to reconstitute average and standard deviation taking BMS algorithm using antiphoneme that near phoneme of CM calculation.

• Journal of Korean Dental Hygiene Science
• /
• v.3 no.2
• /
• pp.39-49
• /
• 2020

The purpose of this study was to provide the basic data needed to plan the effective teaching design of scaling education and to measure the changes in students' confidence in performing scaling according to the number of scaling practices. Scaling education is presented as a core basic hygiene competency of dental hygienists and evaluated as the practical competency of dental hygienists. This study selected 48 third-year students from the department of dental hygiene at S University in Asan, and analyzed data from completed participant surveys. The degree of "confidence in facing the subject in clinical practice," "confidence in applying hand instrument to the subject," and "confidence in applying ultrasonic instruments to the subject" was assessed on a five-point Likert scale. The difference in confidence was analyzed during five practice sessions. Each response was compared using frequency analysis, chi-square test, and repeated measurement ANOVA. Students who complained of a lack of confidence in the 14.6%, 25.0%, and 12.5%, respectively, in face-to-face practice, hand instruments, and ultrasonic instrument application responses. The more the practice was repeated, the more confident students were in all three categories, and the more statistically significant (p<0.001). As the number of scaling practice sessions increased, students' confidence in performing scaling also improved. In particular, the level of self-confidence was higher after the third practice session when compared to the first session. Therefore, it is necessary to design effective courses for teaching scaling practices so that at least three repetitive practice periods can be provided in clinical dental hygiene practices.