• Journal of KIISE
• /
• v.44 no.7
• /
• pp.742-747
• /
• 2017

In this paper, we propose a novel approach to improve the performance of the Convolutional Neural Network(CNN) word embedding model on top of word2vec with the result of performing like doc2vec in conducting a document classification task. The Word Piece Model(WPM) is empirically proven to outperform other tokenization methods such as the phrase unit, a part-of-speech tagger with substantial experimental evidence (classification rate: 79.5%). Further, we conducted an experiment to classify ten categories of news articles written in Korean by feeding words and document vectors generated by an application of WPM to the baseline and the proposed model. From the results of the experiment, we report the model we proposed showed a higher classification rate (89.88%) than its counterpart model (86.89%), achieving a 22.80% improvement. Throughout this research, it is demonstrated that applying doc2vec in the document classification task yields more effective results because doc2vec generates similar document vector representation for documents belonging to the same category.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.291-295
• /
• 2016

본 논문에서는 한국어 기본 유니트 단위로 WPM을 활용한 구글 플레이 스토어 앱의 댓글 감정분석을 수행하였다. 먼저 자동 띄어쓰기 시스템을 적용한 후, 어절단위, 형태소 분석기, WPM을 각각 적용하여 모델을 생성하고, 로지스틱 회귀(Logistic Regression), 소프트맥스 회귀(Softmax Regression), 서포트 벡터머신(Support Vector Machine, SVM)등의 알고리즘을 이용하여 댓글 감정(긍정과 부정)을 비교 분석하였다. 그 결과 어절단위, 형태소 분석기보다 WPM이 최대 25%의 향상된 결과를 얻었다. 또한 분류 과정에서 로지스틱회귀, 소프트맥스 회귀보다는 SVM 성능이 우수했으며, SVM의 기본 파라미터({'kernel':('linear'), 'c':[4]})보다 최적의 파라미터를 적용({'kernel': ('linear','rbf', 'sigmoid', 'poly'), 'C':[0.01, 0.1, 1.4.5]} 하였을 때, 최대 91%의 성능이 나타났다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.291-295
• /
• 2016

본 논문에서는 한국어 기본 유니트 단위로 WPM을 활용한 구글 플레이 스토어 앱의 댓글 감정분석을 수행하였다. 먼저 자동 띄어쓰기 시스템을 적용한 후, 어절단위, 형태소 분석기, WPM을 각각 적용하여 모델을 생성하고, 로지스틱 회귀(Logistic Regression), 소프트맥스 회귀(Softmax Regression), 서포트 벡터머신(Support Vector Machine, SVM)등의 알고리즘을 이용하여 댓글 감정(긍정과 부정)을 비교 분석하였다. 그 결과 어절단위, 형태소 분석기보다 WPM이 최대 25%의 향상된 결과를 얻었다. 또한 분류 과정에서 로지스틱회귀, 소프트맥스 회귀보다는 SVM 성능이 우수했으며, SVM의 기본 파라미터({'kernel':('linear'), 'c':[4]})보다 최적의 파라미터를 적용({'kernel': ('linear','rbf', 'sigmoid', 'poly'), 'C':[0.01, 0.1, 1.4.5]} 하였을 때, 최대 91%의 성능이 나타났다.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.67-71
• /
• 2016

본 논문에서는 word2vec과 doc2vec을 함께 CNN에 적용한 문서 분류 방안을 제안한다. 먼저 어절, 형태소, WPM(Word Piece Model)을 각각 사용하여 생성한 토큰(token)으로 doc2vec을 활용하여 문서를 vector로 표현한 후, 초보적인 문서 분류에 적용한 결과 WPM이 분류율 79.5%가 되어 3가지 방법 중 최고 성능을 보였다. 다음으로 CNN의 입력자질로써 WPM을 이용하여 생성한 토큰을 활용한 word2vec을 범주 10개의 문서 분류에 사용한 실험과 doc2vec을 함께 사용한 실험을 수행하였다. 실험 결과 word2vec만을 활용하였을 때 86.89%의 분류율을 얻었고, doc2vec을 함께 적용한 결과 89.51%의 분류율을 얻었다. 따라서 제안한 모델을 통해서 분류율이 2.62% 향상됨을 확인하였다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.67-71
• /
• 2016

본 논문에서는 word2vec과 doc2vec을 함께 CNN에 적용한 문서 분류 방안을 제안한다. 먼저 어절, 형태소, WPM(Word Piece Model)을 각각 사용하여 생성한 토큰(token)으로 doc2vec을 활용하여 문서를 vector로 표현한 후, 초보적인 문서 분류에 적용한 결과 WPM이 분류율 79.5%가 되어 3가지 방법 중 최고 성능을 보였다. 다음으로 CNN의 입력자질로써 WPM을 이용하여 생성한 토큰을 활용한 word2vec을 범주 10개의 문서 분류에 사용한 실험과 doc2vec을 함께 사용한 실험을 수행하였다. 실험 결과 word2vec만을 활용하였을 때 86.89%의 분류율을 얻었고, doc2vec을 함께 적용한 결과 89.51%의 분류율을 얻었다. 따라서 제안한 모델을 통해서 분류율이 2.62% 향상됨을 확인하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.7
• /
• pp.1380-1386
• /
• 2003

Speech recognition performance is degraded by the variation in vocal tract length among speakers. In this paper, we have used a vocal tract length normalization method wherein the frequency axis of the short-time spectrum associated with a speaker's speech is scaled to minimize the effects of speaker's vocal tract length on the speech recognition performance In order to normalize vocal tract length, we tried several frequency warping functions such as linear and piece-wise linear function. Variable interval piece-wise linear warping function is proposed to effectively model the variation of frequency axis scale due to the large variation of vocal tract length. Experimental results on TIDIGITS connected digits showed the dramatic reduction of word error rates from 2.15% to 0.53% by the proposed vocal tract normalization.

• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.23-34
• /
• 2019

Recently, the crime that utilizes the digital platform is continuously increasing. About 140,000 cases occurred in 2015 and about 150,000 cases occurred in 2016. Therefore, it is considered that there is a limit handling those online crimes by old-fashioned investigation techniques. Investigators' manual online search and cognitive investigation methods those are broadly used today are not enough to proactively cope with rapid changing civil crimes. In addition, the characteristics of the content that is posted to unspecified users of social media makes investigations more difficult. This study suggests the site-based collection and the Open API among the content web collection methods considering the characteristics of the online media where the infringement crimes occur. Since illegal content is published and deleted quickly, and new words and alterations are generated quickly and variously, it is difficult to recognize them quickly by dictionary-based morphological analysis registered manually. In order to solve this problem, we propose a tokenizing method in the existing dictionary-based morphological analysis through WPM (Word Piece Model), which is a data preprocessing method for quick recognizing and responding to illegal contents posting online infringement crimes. In the analysis of data, the optimal precision is verified through the Vote-based ensemble method by utilizing a classification learning model based on supervised learning for the investigation of illegal contents. This study utilizes a sorting algorithm model centering on illegal multilevel business cases to proactively recognize crimes invading the public economy, and presents an empirical study to effectively deal with social data collection and content investigation.

• Journal of Information Technology Services
• /
• v.16 no.4
• /
• pp.149-160
• /
• 2017

Sentiment analysis is the process of determining whether a piece of document, text or conversation is positive, negative, neural or other emotion. Sentiment analysis has been applied for several real-world applications, such as chatbot. In the last five years, the practical use of the chatbot has been prevailing in many field of industry. In the chatbot applications, to recognize the user emotion, sentiment analysis must be performed in advance in order to understand the intent of speakers. The specific emotion is more than describing positive or negative sentences. In light of this context, we propose deep learning models for conducting multi-class sentiment analysis for identifying speaker's emotion which is categorized to be joy, fear, guilt, sad, shame, disgust, and anger. Thus, we develop convolutional neural network (CNN), long short term memory (LSTM), and multi-layer neural network models, as deep neural networks models, for detecting emotion in a sentence. In addition, word embedding process was also applied in our research. In our experiments, we have found that long short term memory (LSTM) model performs best compared to convolutional neural networks and multi-layer neural networks. Moreover, we also show the practical applicability of the deep learning models to the sentiment analysis for chatbot.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.437-440
• /
• 2020

최근 자연어 처리에서는 사전 학습과 전이 학습을 통하여 다양한 과제에 높은 성능 향상을 성취하고 있다. 사전 학습의 대표적 모델로 구글의 BERT가 있으며, 구글에서 제공한 다국어 모델을 포함해 한국의 여러 연구기관과 기업에서 한국어 데이터셋으로 학습한 BERT 모델을 제공하고 있다. 하지만 이런 BERT 모델들은 사전 학습에 사용한 말뭉치의 특성에 따라 이후 전이 학습에서의 성능 차이가 발생한다. 본 연구에서는 소셜미디어에서 나타나는 구어체와 신조어, 특수문자, 이모지 등 일반 사용자들의 문장에 보다 유연하게 대응할 수 있는 한국어 뉴스 댓글 데이터를 통해 학습한 KcBERT를 소개한다. 본 모델은 최소한의 데이터 정제 이후 BERT WordPiece 토크나이저를 학습하고, BERT Base 모델과 BERT Large 모델을 모두 학습하였다. 또한, 학습된 모델을 HuggingFace Model Hub에 공개하였다. KcBERT를 기반으로 전이 학습을 통해 한국어 데이터셋에 적용한 성능을 비교한 결과, 한국어 영화 리뷰 코퍼스(NSMC)에서 최고 성능의 스코어를 얻을 수 있었으며, 여타 데이터셋에서는 기존 한국어 BERT 모델과 비슷한 수준의 성능을 보였다.

• Journal of Korean language education
• /
• v.29 no.3
• /
• pp.325-344
• /
• 2018

The purpose of this study is to summarize PWIM (Picture Words Inductive Model) which is one of learner-centered vocabulary teaching-learning models, and suggest ways to implement them in Korean language education. The pictures that are used in the Korean language education field help visualize the specific shape, color, and texture of the vocabulary that is the learning target; thus, helping beginner learners to recognize the meaning of the sound. Visual material stimulates the intrinsic schema of the learner and not only becomes a 'bridge' connecting the mother tongue and the Korean language, but also reduces difficulty in learning a foreign language because of the ambiguity between meaning and sound in Korean and all languages. PWIM shows commonality with existing learning methods in that it uses visual materials. However, in the past, the teacher-centered learning method has only imitated the teacher because the teacher showed a piece-wise, out-of-life photograph and taught the word. PWIM is a learner-centered learning method that stimulates learners to find vocabulary on their own by presenting visual information reflecting the context. In this paper, PWIM is more suitable for beginner learners who are learning specific concrete vocabulary such as personal identity (mainly objects), residence and environment, daily life, shopping, health, climate, and traffic. The purpose of this study was to develop a method of using PWIM suitable for Korean language learners and teaching procedures. The researchers rearranged the previous research into three steps: brainstorming and word organization, generalization of semantic and morphological rules of extracted words, and application of words. In the case of PWIM, you can go through all three steps at once. Otherwise, it is possible to divide the three steps of PWIM and teach at different times. It is expected that teachers and learners using the PWIM teaching-learning method, which uses realistic visual materials, will enable making an effective class together.