• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.333-337
• /
• 2017

본 논문에서는 개체명 인식과 언어 모델의 다중 학습을 이용한 한국어 개체명 인식 방법을 제안한다. 다중 학습은 1 개의 모델에서 2 개 이상의 작업을 동시에 분석하여 성능 향상을 기대할 수 있는 방법이지만, 이를 적용하기 위해서 말뭉치에 각 작업에 해당하는 태그가 부착되어야 하는 문제가 있다. 본 논문에서는 추가적인 태그 부착 없이 정보를 획득할 수 있는 언어 모델을 개체명 인식 작업과 결합하여 성능 향상을 이루고자 한다. 또한 단순한 형태소 입력의 한계를 극복하기 위해 입력 표상을 자소 및 형태소 품사의 임베딩으로 확장하였다. 기계 학습 방법은 순차적 레이블링에서 높은 성능을 제공하는 Bi-directional LSTM CRF 모델을 사용하였고, 실험 결과 언어 모델이 개체명 인식의 오류를 효과적으로 개선함을 확인하였다.

• School Mathematics
• /
• v.6 no.3
• /
• pp.235-249
• /
• 2004

The purpose of this study was to analyze the error patterns and sentence types in word problems with respect to 1$\frac{3}{4}$$\div$$\frac{1}{2}$ which were made by the pre-service elementary teachers, and to suggest the clues to the education in pre-service. Korean elementary teachers in pre-service misunderstood 'divide with $\frac{1}{2}$' to 'divide to 2' by the Korean linguistic structure. And they showed a new error type of 1$\frac{3}{4}$$\times$2 by the result of calculation. Although they are familiar to 'inclusive algorithm' they are not good at dealing with the fractional divisor. And they are very poor at the 'decision the unit proportion' and the 'inverse of multiplication'. So, it is necessary to teach the meaning of the fractional division as 'decision the unit proportion' and 'inverse of multiplication' and to give several examples with respect to the actual situation and context.

• 재활복지
• /
• v.21 no.3
• /
• pp.187-207
• /
• 2017

The purpose of this study was to explore the characteristics of defining concrete and abstract nouns for the elderly. A total of 382 elderly participated in this study and they were classified into four age groups (i.e., over 55 to under 64, over 65 to under 74, over 75 to under 84, and over 85 year-old group). They performed the word definition task, composed of five concrete and five abstract nouns. The total scores and numbers and ratio of core/supplementary meanings were compared among four elderly groups. The frequency and ratio of error types were also examined. The results showed that all four groups had statistically significant differences in total scores, numbers and ratio of core and supplementary meaning of concrete noun definition task. In addition, abstract noun definition performances revealed group differences except the two groups (over 75 to under 84 and over 85-year-old group). The oldest group showed a sharp increase in error production. The highest ratio of error types were personal experience in over 55 to under 64-year-old group, and over 65 to under 74 year-old groups; and for the target word repetition in over 75 to under 84 year-old group; and no response in over 85 year-old group. In conclusion, both concrete and abstract word defining abilities had age-related deterioration. This decline results from impairment in spreading semantic knowledge within semantic network, which is vulnerable to aging. Characteristics of word definition for elderly can provide basic information to understand various neurolinguistic disorders associated with age.

• Clinical and Experimental Pediatrics
• /
• v.50 no.9
• /
• pp.896-900
• /
• 2007

Purpose : The aim of this study was to investigate the speech problems in benign rolandic epilepsy (BRE) according to the seizure focus in EEG and semiology. Methods : Twenty three patients [right origin (13 patients) or left side (10 patients)] who met the BRE criteria by International League Against Epilepsy (ILAE) were prospectively enrolled. We excluded the patients who had abnormal MRI or showed both side spikes in EEG. Computerized Speech Lab was used to assess the speech characteristics of the patients. Results : The error pattern of laryngeal articulation in BRE was exclusively substitution of stop consonants, these errors showed more frequent in the left group (16.0% vs 25.5%). Voice onset time (VOT) of stop consonants and Total duration (TD) of word in both groups were prolonged than normal control group, especially in left group (P<0.05). The first formant of vowel /o/ and second formant of /e/ were significantly decreased in left group (P<0.05). The right group scored wider on pitch range ($192.9{\pm}54.0Hz$) and energy range in spontaneous speech ($14.2{\pm}6.4db$) than the left group ($233.3{\pm}12.5Hz$, $19.4{\pm}9.3db$, respectively, P>0.05). Duration of counting (5 to 9) in left group slower than right group ($8.6{\pm}1.7$ vs $7.9{\pm}1.8sec$). Conclusion : Our data suggested that interictal spikes and seizures in either centrotemporal sides, especially left side group, may induce speech problems. We recommend the logopedic and phoniatric evaluations of speech in BRE patients.

• Journal of Educational Research in Mathematics
• /
• v.22 no.2
• /
• pp.203-227
• /
• 2012

This was to investigate how the slow learners who specially belonged to low-SES, or North Korean defectors showed their errors in mathematical learning. To conduct the study, two groups for each minority group participated in the study volunteerly during the Winter vacation, in 2011. Based on the preliminary interviews, a total of 15 units were given, focusing on building mathematical basic concepts. As results, they had some errors in common. They both were in lack of understanding of the terminologies and not able to apply the meanings of definitions and theorems to a problem. Because of uncertainty of basic knowledge of mathematics, they easily lost their focus and were apt to make a mistake. Also, they showed clear differences. North Korean defectors were not accustomed to using or understanding the meanings of Chines or English in Korean words in expressing, writing mathematical terminologies and reading data on the context. Technical errors, and misinterpreted errors were found. However, students from the low SES showed that they were familiar with mathematical words and terminologies, but their errors mostly belonged to carelessness because of the lack of mastering mathematical concepts.

• Phonetics and Speech Sciences
• /
• v.13 no.3
• /
• pp.91-101
• /
• 2021

This study aimed to investigate speech motor control in adults who stutter (AWS) by testing whether articulation rate changes with practice and linguistic complexity. Eleven AWS and 11 adults who do not stutter (AWNS) repeated four sentences of different lengths and syntactic complexity [simple-short (SS), simple-long (SL), complex-long (CL), and faulty-long (FL) sentences]. Overall articulation rates of each sentence were measured and compared between groups. Practice effects were evaluated by comparing the articulation rates of the first three, middle four, and last three productions. Overall, the AWS had significantly slower articulation rates than AWNS across the four sentences. The longer sentences showed significantly slower articulation rates than the baseline sentence (SS). The articulation rates of the middle four and the last three productions were significantly faster than those of the first three productions of each sentence in both groups. The articulation rates of the SS, SL, and CL sentences indicated a consistent practice effect. The slower articulation rates of the AWS are consistent with a speech motor limitation. There was no interaction with linguistic complexity or practice, so a slower articulation rate may be a general feature of the speech of AWS. Both AWS and AWNS showed practice effects with faster articulation rates which may reflect a degree of adaptation to the stimuli.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.243-245
• /
• 2016

광학 문자 인식(OCR)을 통해 문서의 글자를 인식할 때 띄어쓰기 오류가 발생한다. 본 논문에서는 이를 해결하기 위해 OCR의 후처리 과정으로 동적 프로그래밍을 이용한 분절(Segmentation) 방식의 띄어쓰기 오류 교정 시스템을 제안한다. 제안하는 시스템의 띄어쓰기 오류 교정 과정은 다음과 같다. 첫째, 띄어쓰기 오류가 있다고 분류된 어절 내의 공백을 모두 제거한다. 둘째, 공백이 제거된 문자열을 동적 프로그래밍을 이용한 분절로 입력 문자열에 대하여 가능한 모든 띄어쓰기 후보들을 찾는다. 셋째, 뉴스 기사 말뭉치와 그 말뭉치에 기반을 둔 띄어쓰기 확률 모델을 참조하여 각 후보의 띄어쓰기 확률을 계산한다. 마지막으로 띄어쓰기 후보들 중 확률이 가장 높은 후보를 교정 결과로 제시한다. 본 논문에서 제안하는 시스템을 이용하여 OCR의 띄어쓰기 오류를 해결할 수 있었다. 향후 띄어쓰기 오류 교정에 필요한 언어 규칙 등을 시스템에 추가한 띄어쓰기 교정시스템을 통하여 OCR의 최종적인 인식률을 향상에 대해 연구할 예정이다.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.167-172
• /
• 1999

구어는 프로그래밍 언어와는 달리 주어진 문장 내에서의 해당 어휘의 뜻(semantic information)을 알고 다른 어휘들과의 연관성 (grammatical information)을 알아야만 적절한 형태소분석이 가능하다. 또한 구어는 방대한 양의 어휘들로 구성되어 있으며 사용하는 사람마다의 다양한 응용과 공식화되기 어려운 수많은 예외들로 운용되기 때문에 단순히 찾아보기표와 오토마타만으로는 형태소분석에 한계가 있다. 이에 본 논문에서는 주어진 어휘집과 그 어휘들로 만들어진 다양한 문장들로부터 구어운용의 근본기제를 스스로 학습해나가는 강화학습중심의 언어모델을 제안하고 실제로 한국어 형태소분석에 적용하여 그 성능과 특성을 파악해보았다. 구어파서의 입력은 음절단위의 발음이며 인간이 문장을 듣거나 보는 것과 동일하게 시간에 따라 순차적으로 입력된다. 파서의 출력 또한 시간에 따라 변화되면서 나타나며 입력된 연속음절을 형태소단위로 분리(segmentation)하고 분류(labeling)한 결과를 나타낸다. 생성인식 언어모델이 기존의 언어모델과 다른 점은 구어 파싱에 있어서 필수적인 미등륵어에 대한 유연성과 앞단의 음성인식기 오류에 적절한 반응(fault tolerance)을 나타내는 것이다.

• Journal of the Korean Society of Laryngology, Phoniatrics and Logopedics
• /
• v.7 no.1
• /
• pp.106-112
• /
• 1996

조음장애(articulation disorders)란 조음기관(예 : 혀, 입술. 치아, 입천장)을 통하여 말소리가 만들어지는 과정에서의 결함을 나타내는데, 이러한 조음장애를 보이는 아동은 '불명료한' 구어를 사용하게 되므로 해서 결국 의사소통의 어려움을 나타내게 된다. 음소를 생략하거나, 다른 음소로 대치하거나, 또는 같은 음소내에서 소리를 왜곡시키는 조음 장애 현상들은 순수 조음 장애 아동들 뿐 아니라, 정신지체, 청각장애, 구개파열, 뇌성마비 등의 장애자들에게서 중복적인 결함으로 나타나기도 한다. 본고에서는 소아의 조음장애를 다루는 임상가가 알아두어야 할 관련 요인들, 조음 오류 평가. 그리고 치료방법에 대하여 고찰하고자 한다.

• The KIPS Transactions:PartA
• /
• v.11A no.1
• /
• pp.97-108
• /
• 2004

This paper proposes to use an algorithmic debugging technique for locating logical errors in Java programs. The algorithmic debugging is a semi-automated debugging technique that builds an execution tree from a source program and locates logical errors, if any, included in the program from the execution tree with top-down method. So, it is very important to build a suitable execution tree from the various programming languages. In this paper we propose the method for building an execution tree iron Java programs and walk through an example. This approach could reduce the number of interactions between a user and a debugging system than the traditional step-wise debugging technique.