• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.178-179
• /
• 2018

Recently, methods of using machine learning in artificial intelligence such as autonomous navigation and speech recognition have been actively studied. Classical image processing methods such as classical boundary detection and pattern recognition have many limitations in order to recognize a specific object or area in a digital image. However, when a machine learning method such as deep-learning is used, Can be obtained. However, basically, a large amount of learning data must be secured for machine learning such as deep-learning. Therefore, it is difficult to apply the machine learning for area classification when the amount of data is very small, such as aerial photographs for environmental analysis. In this study, we apply a transfer-learning technique that can be used when the dataset size of the input image is small and the shape of the input image is not included in the category of the training dataset.

• Journal of Convergence for Information Technology
• /
• v.8 no.2
• /
• pp.141-148
• /
• 2018

The purpose of this paper is to explore the scope of realization of multiple perspectives so that the implementation of the right to be forgotten is more realistic than the ideal information deletion concept. We examined domestic and foreign legal system and technology/service trends, and reflected the classification realization level of service realization, processing type and information characteristics of personal information processor, and legislative/technical factors for multi-level scope analysis. As a result, we have presented a matrix of the range of realization of the right to be forgotten and the scope of diversified regulation by the subject of protection. This study will be extended to the convergence of law and engineering, and will contribute to the prediction of social costs and expansion of the market by identifying the scope of 'deletion rights'.

• Korean Journal of Childcare and Education
• /
• v.10 no.6
• /
• pp.43-56
• /
• 2014

The purpose of this study was to analyze the differences of preschoolers' utterance features by their gender and play situation type. For this purpose, a total of 40 5-year-old children participated in this study. Dyad were participated in each play session during 10 minutes. The play session was videotaped and the videotaped data were transcribed by CBS(2014). The collected data were analyzed by using a independent t-test and paired t-test. The main results are as follows. First, girls' MLU-e, MLU-w, MLU-m were longer than that of boys in a familiar play situation. Second, preschoolers' MLU-w was longer in an unfamiliar play situation than in familiar ones and preschoolers' type-token ratio were higher in an unfamiliar play situation than in familiar ones. Implications for the importance of preschoolers' spontaneous speech are discussed.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.1
• /
• pp.24-32
• /
• 1997

This study aims to describe, at one aspect, the relativity between syntactic structure and prosodic phrasing, and at the other, to establish a suitable phrasing pattern to produce more natural synthetic speech. To get meaningful results, all the word boundaries in the prosodic database were statistically analyzed, and assigned by the proper boundary type. The resulting 10 types of prosodic boundaries were classified into 3 types according to the strength of the breaks, which are zero, minor, and major break respectively. We have found out that the durational information was a main cue to determine the major prosodic boundary. Using the bigram and trigram of syntactic information, we predicted major and minor classification of boundary types. With brigram model, we obtained the correct major break prediction rates of 4.60%, 38.2%, the insertion error rates of 22.8%, 8.4% on each Test-I and Test-II text database respectively. With trigram mode, we also obtained the correct major break prediction rates of 58.3%, 42.8%, the insertion error rates of 30.8%, 42.8%, the insertion error rates of 30.8%, 11.8% on Test-I and Test-II text database respectively.

• Journal of the Korea Convergence Society
• /
• v.11 no.9
• /
• pp.91-97
• /
• 2020

Use of voice-based interfaces that can interact with users is increasing as AI technology develops. To date, however, most of the research on voice-based interfaces has been technical in nature, focused on areas such as improving the accuracy of speech recognition. Thus, the voice of most voice-based interfaces is uniform and does not provide users with differentiated sensibilities. The purpose of this study is to add a emotional factor suitable for the AI interface. To this end, we have derived emotional factors that should be considered in designing voice interface. In addition, we looked at voice trends that differed from country to country. For this study, we conducted interviews with voice industry experts from Finland and Norway, countries that use their own independent languages.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.2
• /
• pp.423-428
• /
• 2017

These days, various and tons of education information are rapidly increasing and spreading due to Internet and smart devices usage. Recently, as e-Learning usage increasing, many instructors and students (learners) need to set a goal to maximize learners' result of education and education system efficiency based on big data analytics via online recorded education historical data. In this paper, the author applied Word2Vec algorithm (neural network algorithm) to find similarity among education words and classification by clustering algorithm in order to objectively recognize and analyze online recorded education historical data. When the author applied the Word2Vec algorithm to education words, related-meaning words can be found, classified and get a similar vector values via learning repetition. In addition, through experimental results, the author proved the part of speech (noun, verb, adjective and adverb) have same shortest distance from the centroid by using clustering algorithm.

• KIEAE Journal
• /
• v.17 no.5
• /
• pp.95-100
• /
• 2017

Purpose: The purpose of this study is to investigate the prior art based on deep learning to objectively calculate the metabolic rate which is the subjective factor for the PMV optimum control and to make a plan for future research based on this study. Methods: For this purpose, the theoretical and technical review and applicability analysis were conducted through various documents and data both in domestic and foreign. Results: As a result of the prior art research, the machine learning model of artificial neural network and deep learning has been used in various fields such as speech recognition, scene recognition, and image restoration. As a representative case, OpenCV Background Subtraction is a technique to separate backgrounds from objects or people. PASCAL VOC and ILSVRC are surveyed as representative technologies that can recognize people, objects, and backgrounds. Based on the results of previous researches on deep learning based on metabolic rate for occupational metabolic rate, it was found out that basic technology applicable to occupational metabolic rate calculation technology to be developed in future researches. It is considered that the study on the development of the activity quantity calculation model with high accuracy will be done.

• Osong Public Health and Research Perspectives
• /
• v.9 no.6
• /
• pp.354-361
• /
• 2018

Objectives: The purpose of this study was to explore the effects of injection laryngoplasty (IL) with hyaluronic acid in patients with vocal fold paralysis (VFP). Methods: A total of 50 patients with VFP participated in this study. Pre- and post-IL assessments were performed, which included analyzing the sustained vowel /a/ phonation, and the patient reading 1 Korean sentence from the "Walk" passage that comprised 25 syllables in 10 words. To investigate the effect of IL on vocal fold function, acoustic analysis (acoustic voice quality index, cepstral peak prominence, maximum phonation time, speaking fundamental frequency) was conducted and auditory-perceptual (grade and overall severity), visual judgment (gap), and self-questionnaire (voice handicap index-10) assessments were performed. Results: The patients with VFP showed statistically significant differences between pre-and post-IL assessments for acoustic and auditory-perception, visual judgment, and self-questionnaire assessments. Conclusion: The patients with VFP showed positive change in vocal fold function between pre- and post-IL measurements. The findings showed that IL with hyaluronic acid is an effective method to improve vocal fold function in patients with VFP.

• The Journal of the Korea Contents Association
• /
• v.19 no.4
• /
• pp.187-196
• /
• 2019

This experimental study examines whether paralinguistic behavior is a deception cue in an interrogation. 92 college students participated in an experiment and were randomly assigned to two conditions. Participant were then asked to take the money or not to take the money according to the condition they were assigned. Then participants had a face-to-face interrogation. During the interrogation, participants' paralinguistic behavior was recorded and used for coding and analysis. Results reveal that participants' paralinguistic behaviors differ depending on question types and deceptive paralinguistic cues are speech speed and fillers for the closed critical question and response latency, response length, and fillers for the open critical question. These findings implicate that part of paralinguistic behavior could be a deception cue and thus these cues might be applicable to deception detection in real world criminal investigations.

• Journal of Digital Convergence
• /
• v.17 no.7
• /
• pp.285-292
• /
• 2019

With the popularization of PC, SNS and IoT, a lot of data is generated and the amount is increasing exponentially. Artificial neural network learning is a topic that attracts attention in many fields in recent years by using huge amounts of data. Artificial neural network learning has shown tremendous potential in speech recognition and image recognition, and is widely applied to a variety of complex areas such as medical diagnosis, artificial intelligence games, and face recognition. The results of artificial neural networks are accurate enough to surpass real human beings. Despite these many advantages, privacy problems still exist in artificial neural network learning. Learning data for artificial neural network learning includes various information including personal sensitive information, so that privacy can be exposed due to malicious attackers. There is a privacy risk that occurs when an attacker interferes with learning and degrades learning or attacks a model that has completed learning. In this paper, we analyze the attack method of the recently proposed neural network model and its privacy protection method.