• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.35-46
• /
• 2023

In this study, we propose a Dual Approach methodology to enhance the accuracy of document classifiers by utilizing both contextual and keyword information. Firstly, contextual information is extracted using Google's BERT, a pre-trained language model known for its outstanding performance in various natural language understanding tasks. Specifically, we employ KoBERT, a pre-trained model on the Korean corpus, to extract contextual information in the form of the CLS token. Secondly, keyword information is generated for each document by encoding the set of keywords into a single vector using an Autoencoder. We applied the proposed approach to 40,130 documents related to healthcare and medicine from the National R&D Projects database of the National Science and Technology Information Service (NTIS). The experimental results demonstrate that the proposed methodology outperforms existing methods that rely solely on document or word information in terms of accuracy for document classification.

• Journal of KIISE:Software and Applications
• /
• v.41 no.7
• /
• pp.492-500
• /
• 2014

This paper presents the performance enhancement of automatic bilingual lexicon extraction by using refinement of pivot-context vectors under the standard pivot-based approach, which is very effective method for less-resource language pairs. In this paper, we gradually improve the performance through two different refinements of pivot-context vectors: One is to filter out unhelpful elements of the pivot-context vectors and to revise the values of the vectors through bidirectional translation probabilities estimated by Anymalign and another one is to remove non-noun elements from the original vectors. In this paper, experiments have been conducted on two different language pairs that are bi-directional Korean-Spanish and Korean-French, respectively. The experimental results have demonstrated that our method for high-frequency words shows at least 48.5% at the top 1 and up to 88.5% at the top 20 and for the low-frequency words at least 43.3% at the top 1 and up to 48.9% at the top 20.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.2 s.40
• /
• pp.159-167
• /
• 2006

Lipreading has been suggested as one of the methods to improve the performance of speech recognition in noisy environment. However, existing methods are developed and implemented only in software. This paper suggests a hardware design for real-time lipreading. For real-time processing and feasible implementation, we decompose the lipreading system into three parts; image acquisition module, feature vector extraction module, and recognition module. Image acquisition module capture input image by using CMOS image sensor. The feature vector extraction module extracts feature vector from the input image by using parallel block matching algorithm. The parallel block matching algorithm is coded and simulated for FPGA circuit. Recognition module uses HMM based recognition algorithm. The recognition algorithm is coded and simulated by using DSP chip. The simulation results show that a real-time lipreading system can be implemented in hardware.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.317-322
• /
• 2010

This paper proposes how to improve performance of the Korean document sentiment-classification system using semantic properties of the sentiment words. A sentiment word means a word with sentiment, and sentiment features are defined by a set of the sentiment words which are important lexical resource for the sentiment classification. Sentiment feature represents different sentiment intensity in general field and in specific domain. In general field, we can estimate the sentiment intensity using a snippet from a search engine, while in specific domain, training data can be used for this estimation. When the sentiment intensity of the sentiment features are estimated, it is called semantic orientation and is used to estimate the sentiment intensity of the sentences in the text documents. After estimating sentiment intensity of the sentences, we apply that to the weights of sentiment features. In this paper, we evaluate our system in three different cases such as general, domain-specific, and general/domain-specific semantic orientation using support vector machine. Our experimental results show the improved performance in all cases, and, especially in general/domain-specific semantic orientation, our proposed method performs 3.1% better than a baseline system indexed by only content words.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.1
• /
• pp.1-8
• /
• 2021

In the political field of news articles, there are polarized and biased characteristics such as conservative and liberal, which is called political bias. We constructed keyword-based dataset to classify bias of news articles. Most embedding researches represent a sentence with sequence of morphemes. In our work, we expect that the number of unknown tokens will be reduced if the sentences are constituted by subwords that are segmented by the language model. We propose a document embedding model with subword tokenization and apply this model to SVM and feedforward neural network structure to classify the political bias. As a result of comparing the performance of the document embedding model with morphological analysis, the document embedding model with subwords showed the highest accuracy at 78.22%. It was confirmed that the number of unknown tokens was reduced by subword tokenization. Using the best performance embedding model in our bias classification task, we extract the keywords based on politicians. The bias of keywords was verified by the average similarity with the vector of politicians from each political tendency.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.267-286
• /
• 2023

The conversational agents such as AI speakers utilize voice conversation for human-computer interaction. Voice recognition errors often occur in conversational situations. Recognition errors in user utterance records can be categorized into two types. The first type is misrecognition errors, where the agent fails to recognize the user's speech entirely. The second type is misinterpretation errors, where the user's speech is recognized and services are provided, but the interpretation differs from the user's intention. Among these, misinterpretation errors require separate error detection as they are recorded as successful service interactions. In this study, various text separation methods were applied to detect misinterpretation. For each of these text separation methods, the similarity of consecutive speech pairs using word embedding and document embedding techniques, which convert words and documents into vectors. This approach goes beyond simple word-based similarity calculation to explore a new method for detecting misinterpretation errors. The research method involved utilizing real user utterance records to train and develop a detection model by applying patterns of misinterpretation error causes. The results revealed that the most significant analysis result was obtained through initial consonant extraction for detecting misinterpretation errors caused by the use of unregistered neologisms. Through comparison with other separation methods, different error types could be observed. This study has two main implications. First, for misinterpretation errors that are difficult to detect due to lack of recognition, the study proposed diverse text separation methods and found a novel method that improved performance remarkably. Second, if this is applied to conversational agents or voice recognition services requiring neologism detection, patterns of errors occurring from the voice recognition stage can be specified. The study proposed and verified that even if not categorized as errors, services can be provided according to user-desired results.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.6
• /
• pp.1531-1540
• /
• 1997

In this paper, we presents automatic Korean continuous speech recognizer which is improved by the speech rate estimation and the compensation methods. Automatic continuous speech recognition is significantly more difficult than isolated word recognition because of coarticulatory effects and variations in speech rate. In order to recognize continuous speech, modeling methods of coarticulatory effects and variations in speech rate are needed. In this paper, the speech rate is measured by change of format, and the compensation is peformed by extracting relatively many feature vectors in fast speech. Coarticulatory effects are modeled by defining 514 Korean diphone set, and ETRI's 445 word DB is used for training speech material. With combining above methods, we implement automatic Korean continuous speech recognizer, which shows improved recognition rate, based on DHMM(Discrete Hidden Markov Model).

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1E
• /
• pp.89-95
• /
• 1994

This paper describes development of a real-time voice dialing system which can recognize around one hundred word vocabularies in speaker independent mode. The voice recognition algorithm in this system is implemented on a DSP board with a telephone interface plugged in an IBM PC AT/486. In the DSP board, procedures for feature extraction, vector quantization(VQ), and end-point detection are performed simultaneously in every 10 msec frame interval to satisfy real-time constraints after detecting the word starting point. In addition, we optimize the VQ codebook size and the end-point detection procedure to reduce recognition time and memory requirement. The demonstration system has been displayed in MOBILAB of the Korean Mobile Telecom at the Taejon EXPO'93.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.191-194
• /
• 2011

문서를 이루는 단어들의 빈도수가 지수법칙(power law)를 따른다는 지프의 법칩(Zipf's law)이 있다. 이러한 단어분포를 고려하여 문서의 토픽을 찾아내는 기계학습법이 디리쉴레 프로세스(Dirichlet process) 이다. 이를 발전시켜서 데이터의 잠재 요인(latent factor)들을 베이즈 확률모델에 기반한 샘플링 바탕으로 찾는 방법이 인디언 뷔페 과정(Indian buffet process) 이다. 우리는 25가지의 특징(feature)들에 대한 점수(rating)들이 볼드(blood oxygen dependent level) 신호와 함께 주어지는 PBAIC 2007 데이터에 주성분 분석법(principal component analysis)를 적용했다. PBAIC 2007 데이터는 비디오 게임을 수행하며 기능적뇌영상(functional magnetic resonance imaging, fMRI) 촬영을 하여 얻어진 공개데이터이다. 우리의 연구에서는 주성분 분석법을 이용하여 10개의 독립 성분(independent component)들을 찾았다. 그리고 1.75초 마다 촬영된 BOLD 신호와 10개의 고유벡터(eigenvector)들간의 내적을 취하여 가중치(weight)를 구하였다. 성분들의 가중치를 낮은 순서로 정렬함으로써 각 시간마다 주도적으로 영향을 미치는 성분들을 알아낼 수 있었다.

• Journal of KIISE:Software and Applications
• /
• v.29 no.10
• /
• pp.747-754
• /
• 2002

In this paper, We propose a document thematic words extraction by using principal component analysis(PCA) which is one of the multivariate statistical methods. The proposed PCA model understands the flow of words in the document by using an eigenvalue and an eigenvector, and extracts thematic words. The proposed model is estimated by applying to document summarization. Experimental results using newspaper articles show that the proposed model is superior to the model using either word frequency or information retrieval thesaurus. We expect that the Proposed model can be applied to information retrieval , information extraction and document summarization.