• Journal of the Korean Society for Library and Information Science
• /
• v.45 no.2
• /
• pp.251-275
• /
• 2011

In this paper, we propose a novel kernel called a semantic parse tree kernel that extends the parse tree kernel previously studied to extract protein-protein interactions(PPIs) and shown prominent results. Among the drawbacks of the existing parse tree kernel is that it could degenerate the overall performance of PPI extraction because the kernel function may produce lower kernel values of two sentences than the actual analogy between them due to the simple comparison mechanisms handling only the superficial aspects of the constituting words. The new kernel can compute the lexical semantic similarity as well as the syntactic analogy between two parse trees of target sentences. In order to calculate the lexical semantic similarity, it incorporates context-based word sense disambiguation producing synsets in WordNet as its outputs, which, in turn, can be transformed into more general ones. In experiments, we introduced two new parameters: tree kernel decay factors, and degrees of abstracting lexical concepts which can accelerate the optimization of PPI extraction performance in addition to the conventional SVM's regularization factor. Through these multi-strategic experiments, we confirmed the pivotal role of the newly applied parameters. Additionally, the experimental results showed that semantic parse tree kernel is superior to the conventional kernels especially in the PPI classification tasks.

• The Journal of Korean Association of Computer Education
• /
• v.17 no.6
• /
• pp.115-122
• /
• 2014

One of the problems with current youth generations is increasing rate of violence and suicide in their school lives, and this study aims at the design of a sentiment analysis system to prevent suicide by uising big data process. The main issues of the design are economical implementation, easy and fast processing for the users, so, the open source Hadoop system with MapReduce algorithm is used on the HDFS(Hadoop Distributed File System) for the experimentation. This study uses word count method to do the sentiment analysis with informal data on some sns communications concerning a kinds of violent words, in terms of text mining to avoid some expensive and complex statistical analysis methods.

• Journal of Gifted/Talented Education
• /
• v.26 no.4
• /
• pp.699-719
• /
• 2016

Alternate pictures that are proven to be equivalent are in high demand to assess creative thinking and language skills. This study aimed to investigate the equivalence of three pictures ('Dog owners,' 'Lost Dog,' and 'Overslept') recently developed for use in a creative writing task. Middle school students (N=183) wrote a story in English based on one of the three prompts distributed randomly. Four writing features (fluency, syntactic complexity, lexical diversity, and temporality) were analyzed with Coh-Metrix and MANCOVA. The three prompts were largely equivalent in their capacity to detect differences among writers in all the features of writing. The difficulty levels of the three prompts, however, were not necessarily the same. Two prompts, Dog Owners and Lost Dog, were verified as equivalent prompts, and therefore, they are recommended as alternate forms to assess creative language skills in repeated measurements. The Overslept prompt had greater facility in eliciting diverse words and more temporal connectives in composing stories. The differential difficulty shown among the prompts suggests that the validity of using different picture versions in repeated assessment remains questionable unless those versions undergo equivalence verification.

• Journal of Korean language education
• /
• v.29 no.4
• /
• pp.255-294
• /
• 2018

The Buckwheat Season, evaluated as the best of Lee Hyo-seok's literature, is one of the short stories that represent Korean literature. However, vivid literary expressions such as lyrical and beautiful depictions, figurative expressions and dialects, which show the Korean beauty, rather make learners have difficulty and become a factor that fails in reading comprehension. Thus, it is necessary to revise and present the text modified for the learners' language level. The methods of revising a literary text include the revision of linguistic elements such as cryptic vocabulary or sentence structure and the revision of the composition of the text, e.g. suggestion of characters or plot, or insertion of illustration. The methods of revising the language of the text can be divided into methods of simplification and detailing. However, in the process of revising the text, many depend on the adapter's subjective perception, not revising it with objective criteria. This paper revised the text, utilizing by the Academy of Korean Studies, , and the by the National Institute of Korean Language to secure objectivity in revising the text.

• Journal of KIISE:Databases
• /
• v.34 no.6
• /
• pp.554-563
• /
• 2007

The e-Marketplace is growing rapidly and providing a more complex relationship between providers and consumers. In recent years, e-Marketplace integration or cooperation issues have become an important issue in e-Business. The e-Catalog is a key factor in e-Business, which means an e-Catalog System needs to contain more large data and requires a more efficient retrieval system. This paper focuses on designing an efficient retrieval system for very large e-Catalogs of large e-Marketplaces. For this reason, a new similarity measure for e-Catalog retrieval based on semantic relationships was proposed. Our achievement is this: first, a new e-Catalog data model based on semantic relationships was designed. Second, the model was extended by considering lexical features (Especially, focus on Korean). Third, the factors affecting similarity with the model was defined. Fourth, from the factors, we finally defined a new similarity measure, realized the system and verified it through experimentation.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.213-218
• /
• 2019

기계독해는 자연어로 표현된 질문과 단락이 주어졌을 때, 해당 단락 내에 표현된 정답을 찾는 태스크이다. 최근 기계독해 태스크도 다른 자연어처리 태스크와 유사하게 BERT, XLNet, RoBERTa와 같이 사전에 학습한 언어모델을 이용하고 질문과 단락이 입력되었을 경우 정답의 경계를 추가 학습(fine-tuning)하는 방법이 우수한 성능을 보이고 있으며, 특히 KorQuAD v1.0 데이터셋에서 학습 및 평가하였을 경우 94% F1 이상의 높은 성능을 보이고 있다. 본 논문에서는 현재 최고 수준의 기계독해 기술이 학습셋과 유사한 평가셋이 아닌 일반적인 질문과 단락 쌍에 대해서 가지는 일반화 능력을 평가하고자 한다. 이를 위하여 첫번째로 한국어에 대해서 공개된 KorQuAD v1.0 데이터셋과 NIA v2017 데이터셋, 그리고 엑소브레인 과제에서 구축한 엑소브레인 v2018 데이터셋을 이용하여 데이터셋 간의 교차 평가를 수행하였다. 교차 평가결과, 각 데이터셋의 정답의 길이, 질문과 단락 사이의 오버랩 비율과 같은 데이터셋 통계와 일반화 성능이 서로 관련이 있음을 확인하였다. 다음으로 KorBERT 사전 학습 언어모델과 학습 가능한 기계독해 데이터 셋 21만 건 전체를 이용하여 학습한 기계독해 모델에 대해 블라인드 평가셋 평가를 수행하였다. 블라인드 평가로 일반분야에서 학습한 기계독해 모델의 법률분야 평가셋에서의 일반화 성능을 평가하고, 정답 단락을 읽고 질문을 생성하지 않고 질문을 먼저 생성한 후 정답 단락을 검색한 평가셋에서의 기계독해 성능을 평가하였다. 블라인드 평가 결과, 사전 학습 언어 모델을 사용하지 않은 기계독해 모델 대비 사전 학습 언어 모델을 사용하는 모델이 큰 폭의 일반화 성능을 보였으나, 정답의 길이가 길고 질문과 단락 사이 어휘 오버랩 비율이 낮은 평가셋에서는 아직 80%이하의 성능을 보임을 확인하였다. 본 논문의 실험 결과 기계 독해 태스크는 특성 상 질문과 정답 사이의 어휘 오버랩 및 정답의 길이에 따라 난이도 및 일반화 성능 차이가 발생함을 확인하였고, 일반적인 질문과 단락을 대상으로 하는 기계독해 모델 개발을 위해서는 다양한 유형의 평가셋에서 일반화 평가가 필요함을 확인하였다.

• Journal of The Korean Association of Information Education
• /
• v.20 no.2
• /
• pp.197-206
• /
• 2016

A fill-in-the-blank with choices are widely used in classrooms in order to check whether students' understand what is being taught. Although there have been proposed many algorithms for generating this type of questions, most of them focus on preparing sentences with blanks rather than generating multiple choices. In this paper, we propose a novel algorithm for generating multiple choices, given a sentence with a blank. Because the algorithm is based on a statistical language model, we can generate relatively unbiased result and adjust the level of difficulty with ease. The experimental results show that our approach automatically produces similar multiple-choices to those of the exam writers.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1452-1458
• /
• 2015

Korean Morphological Analysis is a difficult process. Because Korean is an agglutinative language, one of the most important processes in Morphological Analysis is Morpheme Recovery. There are some methods using Heuristic rules and Pre-Analyzed Partial Words that were examined for this process. These methods have performance limits as a result of not using contextual information. In this study, we built a Korean morpheme recovery system using deep learning, and this system used word embedding for the utilization of contextual information. In '들/VV' and '듣/VV' morpheme recovery, the system showed 97.97% accuracy, a better performance than with SVM(Support Vector Machine) which showed 96.22% accuracy.

• Annual Conference on Human and Language Technology
• /
• 2002.10e
• /
• pp.129-136
• /
• 2002

고성능의 질의응답 시스템을 구현하기 위해서는 사용자의 질의 유형의 난이도에 관계없이 의도를 파악할 수 있는 질의유형 분류기가 필요하다. 본 논문에서는 문서 범주화 기법을 이용한 질의 유형 분류기를 제안한다. 본 논문에서 제안하는 질의 유형 분류기의 분류 과정은 다음과 같다. 우선, 사용자 질의에 포함된 어휘, 품사, 의미표지와 같은 다양한 정보를 이용하여 사용자 질의로부터 자질들을 추출한다. 이 과정에서 질의의 구문 특성을 반영하기 위해서 슬라이딩 윈도 기법을 이용한다. 또한, 다량의 자질들 중에서 유용한 것들만을 선택하기 위해서 카이 제곱 통계량을 이용한다. 추출된 자질들은 벡터 공간 모델로 표현되고, 문서 범주화 기법 중 하나인 지지 벡터 기계(support vector machine, SVM)는 이 정보들을 이용하여 질의 유형을 분류한다. 본 논문에서 제안하는 시스템은 질의 유형 분류 문제에지지 벡터 기계를 이용한 자동문서 범주화 기법을 도입하여 86.4%의 높은 분류 정확도를 보였다. 또한 질의 유형 분류기를 통계적 방법으로 구축함으로써 lexico-syntactic 패턴과 같은 규칙을 기술하는 수작업을 배제할 수 있으며, 응용 영역의 변화에 대해서도 안정적인 처리와 빠른 이식성을 보장한다.

• The Journal of the Korea Contents Association
• /
• v.21 no.10
• /
• pp.576-593
• /
• 2021

In this study, in order to investigate the factors that affect the percentage of correct answers and the degree of discrimination of the TOEIC test, a regression analysis was performed using corpus indicators that influence correct answer rate and the degree of discrimination for each part derived from the item analysis. The basic calculation word_length, consistency index LSA_overlap_adjacent_sentences, lexical diversity MTLD_VOCD, conjunction All_logical_causal_connectives_incidence, situational model casual_particles_causal_verbs_Ratio, syntactic complexity Left_embeddedness, and syntactic pattern density Infinitive_density were found to have negative effects. These factors that lower the correct answer rate can be utilized when setting learning goals. Vocabulary diversity index MTLD_VOCD, conjunction Additive_connectives_incidence, syntactic pattern density Infinitive_density, and lexical information person1_2_pronoun_incidence were found to have a positive effect. Factors influencing the increase in discrimination may provide important information for developing a learning program.