• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.1030-1032
• /
• 2019

형태소 분석을 통하여 한국어 문장을 형태소 단위의 임베딩 및 학습 관련 연구가 되었으나 최근 비정형적인 텍스트 데이터의 증가에 따라 음소 단위의 임베딩을 통한 신경망 학습에 대한 요구가 높아지고 있다. 본 논문은 비정형적인 텍스트 감성 분석 성능 향상을 위해 음소 단위의 토큰을 생성하고 이를 CNN 모형을 기반으로 다차원 임베딩을 수행하고 감성분석을 위하여 양방향 순환신경망 모델을 사용하여 유튜브의 비정형 텍스트를 학습시켰다. 그 결과 텍스트의 긍정 부정 판별에 있어 90%의 정확도를 보였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2004.05b
• /
• pp.231-235
• /
• 2004

본 연구의 목적은 물 순환의 한 부분을 이루는 지하수에 대한 순환과정을 파악하는 것이다. 이를 위해 금강유역의 10개 지하수 소유역에 대해 지하수 함양량 및 유출량을 추정하였으려, 강수량, 하천유출량 및 증발산량과의 비교를 통해 거시적인 지하수 물 순환 분석을 수행하였다. 각 소유역에서의 지하수 함양량 및 유출량은 하천의 일유출 수문곡선으로부터 추정하는 방법을 이용하였으며, 토양수분 저류구조를 갖는 탱크모형을 이용하여 자 소유역별 하천유출량을 생성한 후 생성된 유출량을 분석하여 함양량 및 유출량에 대한 각 소유역별 특성치를 분석하였다. 분석결과 지하수 함양은 봄철부터 여름철까지 지속적으로 상승하는 형태를 나타내고 있었으며, 가을철에는 매우 작은 함양량을 나타내고 있었다. 지하수 유출의 경우에는 다른 계절에 비해 여름철에 상대적으로 많은 양의 유출을 보이고 있었으며, 나머지 다른 계절에는 상대적으로 유사한 크기의 유출을 보이고 있었다. 또한 모형을 통해 분석된 연간 지하수 함양량과 유출량은 매우 유사한 값을 나타내고 있었으며, 총 강수량과의 비교에서도 과거 연구결과와 유사한 길과를 나타내고 있었다. 따라서 본 연구에서 제시된 유역 내 지하수 물 순환의 정량적 해석은 지표수-지하수 연계 운영의 거시적 해석에 필요한 기본 정보를 줄 수 있을 것으로 판단된다.

• Journal of the Korean Society for Library and Information Science
• /
• v.46 no.1
• /
• pp.163-181
• /
• 2012

We investigated how useful video tags were in summarizing video speech and how valuable positional information was for speech summarization. Furthermore, we examined the similarity among sentences selected for a speech summary to reduce its redundancy. Based on such analysis results, we then designed and evaluated a method for automatically summarizing speech transcripts using a modified Maximum Marginal Relevance model. This model did not only reduce redundancy but it also enabled the use of social tags, title words, and sentence positional information. Finally, we compared the proposed method to the Extractor system in which key sentences of a video speech were chosen using the frequency and location information of speech content words. Results showed that the precision and recall rates of the proposed method were higher than those of the Extractor system, although there was no significant difference in the recall rates.

• Annual Conference on Human and Language Technology
• /
• 2015.10a
• /
• pp.133-136
• /
• 2015

문서 교정기에서 문맥의존 철자오류를 교정하는 방법은 크게 규칙을 이용한 방법과 통계 정보를 이용한 방법으로 나뉜다. 한국어와 달리 영어는 오래전부터 통계 모형에 기반을 둔 문맥의존 철자오류 교정 연구가 활발히 이루어졌다. 그러나 대부분 연구가 문맥의존 철자오류 교정 문제를 특정 어휘 쌍을 이용한 분류 문제로 간주하기 때문에 실제 응용에는 한계가 있다. 또한, 대규모 말뭉치에서 추출한 통계 정보를 이용하지만, 통계 정보 자체에 오류가 있을 경우를 고려하지 않았다. 본 논문에서는 텍스트에 포함된 모든 단어에 대하여 문맥의존 철자오류 여부를 판단하고, 해당 단어가 오류일 경우 대치어를 제시하는 영어 문맥의존 철자오류 교정 기법을 제안한다. 또한, 통계 정보의 오류가 문맥의존 철자오류 교정에 미치는 영향과 오류 발생률의 변화가 철자오류 검색과 교정의 정확도와 재현율에 미치는 영향을 분석한다. 구글 웹데이터에서 추출한 통계 정보를 바탕으로 통계 모형을 구성하고 평가를 위해 브라운 말뭉치에서 무작위로 2,000문장을 추출하여 무작위로 문맥의존 철자오류를 생성하였다. 실험결과, 문맥의존 철자오류 검색의 정확도와 재현율은 각각 98.72%, 95.79%였으며, 문맥의존 철자오류 교정의 정확도와 재현률은 각각 71.94%, 69.81%였다.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.3
• /
• pp.13-23
• /
• 2022

In this study we leveraged a probing method to evaluate how a pre-trained L2 LSTM language model represents sentences with relative and coordinate clauses. The probing experiment employed adapted models based on the pre-trained L2 language models to trace the syntactic properties of sentence embedding vector representations. The dataset for probing was automatically generated using several templates related to different sentence structures. To classify the syntactic properties of sentences for each probing task, we measured the adaptation effects of the language models using syntactic priming. We performed linear mixed-effects model analyses to analyze the relation between adaptation effects in a complex statistical manner and reveal how the L2 language models represent syntactic features for English sentences. When the L2 language models were compared with the baseline L1 Gulordava language models, the analogous results were found for each probing task. In addition, it was confirmed that the L2 language models contain syntactic features of relative and coordinate clauses hierarchically in the sentence embedding representations.

• School Mathematics
• /
• v.5 no.3
• /
• pp.361-384
• /
• 2003

The purpose of this study was to investigate students' problem solving process based on the model of IDEAL if they learn to solve word problems of simultaneous linear equations through structure-representation instruction. The problem solving model of IDEAL is followed by stages; identifying problems(I), defining problems(D), exploring alternative approaches(E), acting on a plan(A). 160 second-grade students of middle schools participated in a study was classified into those of (a) a control group receiving no explicit instruction of structure-representation in word problem solving, and (b) a group receiving structure-representation instruction followed by IDEAL. As a result of this study, a structure-representation instruction improved word-problem solving performance and the students taught by the structure-representation approach discriminate more sharply equivalent problem, isomorphic problem and similar problem than the students of a control group. Also, students of the group instructed by structure-representation approach have less errors in understanding contexts and using data, in transferring mathematical symbol from internal learning relation of word problem and in setting up an equation than the students of a control group. Especially, this study shows that the model of direct transformation and the model of structure-schema in students' problem solving process of I and D stages.

• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.11-29
• /
• 2022

Pre-trained language models are the most important and widely used tools in natural language processing tasks. Since those have been pre-trained for a large amount of corpus, high performance can be expected even with fine-tuning learning using a small number of data. Since the elements necessary for implementation, such as a pre-trained tokenizer and a deep learning model including pre-trained weights, are distributed together, the cost and period of natural language processing has been greatly reduced. Transformer variants are the most representative pre-trained language models that provide these advantages. Those are being actively used in other fields such as computer vision and audio applications. In order to make it easier for researchers to understand the pre-trained language model and apply it to natural language processing tasks, this paper describes the definition of the language model and the pre-learning language model, and discusses the development process of the pre-trained language model and especially representative Transformer variants.

• Journal of The Korean Association of Information Education
• /
• v.14 no.1
• /
• pp.1-12
• /
• 2010

For hearing-impaired students, it is known that reading ability is the most important means of communication. In the meanwhile, with recent development of wireless communication technologies, mobile devices are used in various education fields. The purpose of this study is to design and implement a mobile system to improve reading ability of hearing-impaired students. For this purpose, "Question Generation Strategy", known as one of the effective methods for improving reading ability, is adopted to make study contents. Also, 2Bi (Bilingual-Bicultural) Approach Model, an attractive model for improving reading ability of hearing-impaired students, is used. Characteristics of the proposed mobile system are as follows. First, the system is developed to let students learn written language usage through repetition and difference of two organically-related curriculums for hearing-impaired students. Second, study contents are made to increase sentence understanding ability using an activity that is to let students read articles, make questions and answer questions for themselves. Third, the proposed system is designed and implemented to allow students to choose study contents individually anytime anywhere depending on their study levels.

• Journal of Gifted/Talented Education
• /
• v.24 no.3
• /
• pp.421-444
• /
• 2014

In STEAM education, contents that has been extracted from a variety of areas, so it can work closely and systematically. Therefore STEAM education requires systems thinking that can be grasped effectively these different disciplines. The purposes of this study are to develop a STEAM program based on systems thinking, and apply the program to the secondary science gifted student in order to investigate the educational effect. A model of the Program developed from previous research and theoretical contents of systems thinking and STEAM. A draft of the STEAM program was developed on the theme of "rocket". A total of 113 students was participated in this study. 100 seventh and 13 eighth graders were enrolled at seigy. A single group pre-post test paired t-test was conducted on them in systems thinking skills. Result of applying the program to the students as follows. The systems thinking ability was improved after the application of the program. 'Mental Model', 'Personal Skill', 'Team Learning', and 'System Analysis', 'Shared Vision' emerged for both improved significantly. In conclusion, the STEAM program based on system thinking improves students' systems thinking skills. This program of results can be helpful in cultivate human resources with the problem solving ability based on system thinking and STEAM literacy by used in public education curriculum.