• Journal of the Korean Society for Library and Information Science
• /
• v.39 no.3
• /
• pp.93-103
• /
• 2005

Users of searching Internet English dictionary sometimes do not know the correct spelling of the word in mind, but remember only its pronunciation. In order to help these users, we propose a method to retrieve English words effectively with a spoken word transliteration that is a Korean transliteration of English word pronunciation. We develop KONIX codes and transform a spoken word transliteration and English words into them. We then calculate the phonetic similarity between KONIX codes using edit distance and 2-gram methods. Experimental results show that the proposed method is very effective for retrieving English words with a spoken word transliteration.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.297-304
• /
• 2010

Clustering method which based on sentence type or document genre is a technique used to improve translation quality of SMT(statistical machine translation) by domain-specific translation. But there is no previous research using sentence type and document genre information simultaneously. In this paper, we suggest an integrated clustering method that classifying sentence type by syntactic structure similarity and document genre by word similarity information. We interpolated domain-specific models from clusters with general models to improve translation quality of SMT system. Kernel function and cosine measures are applied to calculate structural similarity and word similarity. With these similarities, we used machine learning algorithms similar to K-means to clustering. In Japanese-English patent translation corpus, we got 2.5% point relative improvements of translation quality at optimal case.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.441-443
• /
• 2021

Sentence or text similarity is a measure of the degree of similarity between two sentences. Techniques for measuring text similarity include Jacquard similarity, cosine similarity, Euclidean similarity, and Manhattan similarity. Currently, the cosine similarity technique is most often used, but since this is an analysis according to the occurrence or frequency of a word in a sentence, the analysis on the semantic relationship is insufficient. Therefore, we try to improve the efficiency of analysis on the similarity of sentences by giving relations between words using ontology and including semantic similarity when extracting words that are commonly included in two sentences.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.282-289
• /
• 2008

Although there were some technological developments in improving the collaborative filtering, they have yet to fully reflect the actual relation of the items. In this paper, we propose the recommendation system using associative web document classification by word frequency and ${\alpha}$-cut to address the short comings of the collaborative filtering. The proposed method extracts words from web documents through the morpheme analysis and accumulates the weight of term frequency. It makes associative rules and applies the weight of term frequency to its confidence by using Apriori algorithm. And it calculates the similarity among the words using the hypergraph partition. Lastly, it classifies related web document by using ${\alpha}$-cut and calculates similarity by using adjusted cosine similarity. The results show that the proposed method significantly outperforms the existing methods.

• Annual Conference on Human and Language Technology
• /
• 2015.10a
• /
• pp.165-168
• /
• 2015

기계학습 기반의 자연어처리 모듈에서 중요한 단계 중 하나는 모듈의 입력으로 단어를 표현하는 것이다. 벡터의 사이즈가 크고, 단어 간의 유사성의 개념이 존재하지 않는 One-hot 형태와 대조적으로 유사성을 표현하기 위해서 단어를 벡터로 표현하는 단어 표현 (word representation/embedding) 생성 작업은 자연어 처리 작업의 기계학습 모델의 성능을 개선하고, 몇몇 자연어 처리 분야의 모델에서 성능 향상을 보여 주어 많은 관심을 받고 있다. 본 논문에서는 Word2Vec, CCA, 그리고 GloVe를 사용하여 106,552개의 PubMed의 바이오메디컬 논문의 요약으로 구축된 말뭉치 카테고리의 각 단어 표현 모델의 카테고리 분류 능력을 확인한다. 세부적으로 나눈 카테고리에는 질병의 이름, 질병 증상, 그리고 난소암 마커가 있다. 분류 능력을 확인하기 위해 t-SNE를 이용하여 2차원으로 단어 표현 결과를 맵핑하여 가시화 한다.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.337-339
• /
• 2016

본 논문에서는 한국어 개체명 인식의 성능 향상을 위하여 워드 임베딩을 활용할 수 있는 방법에 대하여 기술한다. 워드 임베딩이란 문장의 단어의 공기정보를 바탕으로 그 단어의 의미를 벡터로 표현하는 분산표현이다. 이러한 분산 표현은 단어 간의 유의미한 정도를 계산하는데 유용하다. 본 논문에서는 이러한 워드 임베딩을 통하여 단어 벡터들의 코사인 유사도를 통한 개체명 사전 자가 학습 및 매칭 방법을 적용하고, 그 실험 결과를 보고한다.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.60-63
• /
• 2017

자연어처리에 딥 러닝을 적용하기 위해 사용되는 Word embedding은 단어를 벡터 공간상에 표현하는 것으로 차원축소 효과와 더불어 유사한 의미의 단어는 유사한 벡터 값을 갖는다는 장점이 있다. 이러한 word embedding은 대용량 코퍼스를 학습해야 좋은 성능을 얻을 수 있기 때문에 기존에 많이 사용되던 word2vec 모델은 대용량 코퍼스 학습을 위해 모델을 단순화 하여 주로 단어의 등장 비율에 중점적으로 맞추어 학습하게 되어 단어의 위치 정보를 이용하지 않는다는 단점이 있다. 본 논문에서는 기존의 word embedding 학습 모델을 단어의 위치정보를 이용하여 학습 할 수 있도록 수정하였다. 실험 결과 단어의 위치정보를 이용하여 word embedding을 학습 하였을 경우 word-analogy의 syntactic 성능이 크게 향상되며 어순이 바뀔 수 있는 한국어에서 특히 큰 효과를 보였다.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.60-63
• /
• 2017

자연어처리에 딥 러닝을 적용하기 위해 사용되는 Word embedding은 단어를 벡터 공간상에 표현하는 것으로 차원축소 효과와 더불어 유사한 의미의 단어는 유사한 벡터 값을 갖는다는 장점이 있다. 이러한 word embedding은 대용량 코퍼스를 학습해야 좋은 성능을 얻을 수 있기 때문에 기존에 많이 사용되던 word2vec 모델은 대용량 코퍼스 학습을 위해 모델을 단순화 하여 주로 단어의 등장 비율에 중점적으로 맞추어 학습하게 되어 단어의 위치 정보를 이용하지 않는다는 단점이 있다. 본 논문에서는 기존의 word embedding 학습 모델을 단어의 위치정보를 이용하여 학습 할 수 있도록 수정하였다. 실험 결과 단어의 위치정보를 이용하여 word embedding을 학습 하였을 경우 word-analogy의 syntactic 성능이 크게 향상되며 어순이 바뀔 수 있는 한국어에서 특히 큰 효과를 보였다.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.3
• /
• pp.3-9
• /
• 1999

This paper presents the study on the rejection capability based on anti-phone modeling for vocabulary independent speech recognition system. The rejection system detects and rejects out-of-vocabulary words which were not included in candidate words which are defined while the speech recognizer is made. The rejection system can be classified into two categories by their implementation methods, keyword spotting method and utterance verification method. The keyword spotting method uses an extra filler model as a candidate word as well as keyword models. The utterance verification method uses the anti-models for each phoneme for the calculation of confidence score after it has constructed the anti-models for all phonemes. We implemented an utterance verification algorithm which can be used for vocabulary independent speech recognizer. We also compared three kinds of means for the calculation of confidence score, and found out that the geometric mean had shown the best result. For the normalization of confidence score, usually Sigmoid function is used. On using it, we compared the effect of the weight constant for Sigmoid function and determined the optimal value. And we compared the effects of the size of cohort set, the results showed that the larger set gave the better results. And finally we found out optimal confidence score threshold value. In case of using the threshold value, the overall recognition rate including rejection errors was about 76%. This results are going to be adapted for stock information system based on speech recognizer which is currently provided as an experimental service by Korea Telecom.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.13 no.2
• /
• pp.102-110
• /
• 2017

Current word sense disambiguation techniques employ various machine learning-based methods. Various approaches have been proposed to address this problem, including the knowledge base approach. This approach defines the sense of an ambiguous word in accordance with knowledge base information with no training corpus. In unsupervised learning techniques that use a knowledge base approach, graph-based and similarity-based methods have been the main research areas. The graph-based method has the advantage of constructing a semantic graph that delineates all paths between different senses that an ambiguous word may have. However, unnecessary semantic paths may be introduced, thereby increasing the risk of errors. To solve this problem and construct a fine-grained graph, in this paper, we propose a model that iteratively constructs the graph while eliminating unnecessary nodes and edges, i.e., senses and semantic paths. The hybrid similarity estimation model was applied to estimate a more accurate sense in the constructed semantic graph. Because the proposed model uses BabelNet, a multilingual lexical knowledge base, the model is not limited to a specific language.