• Journal of Information Technology Applications and Management
• /
• 제22권3호
• /
• pp.83-103
• /
• 2015

The rapid development of internet technologies and social media over the last few years has generated a huge amount of unstructured text data, which contains a great deal of valuable information and issues. Therefore, text mining-extracting meaningful information from unstructured text data-has gained attention from many researchers in various fields. Topic analysis is a text mining application that is used to determine the main issues in a large volume of text documents. However, it is difficult to identify related issues or meaningful insights as the number of issues derived through topic analysis is too large. Furthermore, traditional issue-clustering methods can only be performed based on the co-occurrence frequency of issue keywords in many documents. Therefore, an association between issues that have a low co-occurrence frequency cannot be recognized using traditional issue-clustering methods, even if those issues are strongly related in other perspectives. Therefore, in this research, a methodology to reorganize social issues from a research and development (R&D) perspective using social network analysis is proposed. Using an R&D perspective lexicon, issues that consistently share the same R&D keywords can be further identified through social network analysis. In this study, the R&D keywords that are associated with a particular issue imply the key technology elements that are needed to solve a particular issue. Issue clustering can then be performed based on the analysis results. Furthermore, the relationship between issues that share the same R&D keywords can be reorganized more systematically, by grouping them into clusters according to the R&D perspective lexicon. We expect that our methodology will contribute to establishing efficient R&D investment policies at the national level by enhancing the reusability of R&D knowledge, based on issue clustering using the R&D perspective lexicon. In addition, business companies could also utilize the results by aligning the R&D with their business strategy plans, to help companies develop innovative products and new technologies that sustain innovative business models.

• 한국안전학회지
• /
• 제33권3호
• /
• pp.58-64
• /
• 2018

Recently, a lot of accident report documents have accumulated in almost all of industries, including critical information of accidents. Accordingly, text data contained in accident report documents are considered useful information for understanding accident processes. However, there has been a lack of systematic approaches to analyzing accident report documents. In this respect, this paper aims at proposing text analytics approach to extracting critical information on accident processes. To be specific, major causes of the accident occurrence are classified based on text information contained in accident report documents by using both textmining and latent Dirichlet allocation (LDA) algorithms. The textmining algorithm is used to structure the document-term matrix and the LDA algorithm is applied to extract latent topics included in a lot of accident report documents. We extract ten topics of accidents as accident types and related keywords of accidents with respect to each accident type. The cause-and-effect diagram is then depicted as a tool for navigating processes of the accident occurrence by structuring causes extracted from LDA. Further, the trends of accidents are identified to explore patterns of accident occurrence in each of types. Three patterns of increasing to decreasing, decreasing to increasing, or only increasing are presented in the case of a chemical plant. The proposed approach helps safety managers systematically supervise the causes and processes of accidents through analysis of text information contained in accident report documents.

• 응용통계연구
• /
• 제35권4호
• /
• pp.469-484
• /
• 2022

텍스트 데이터는 일반적으로 많은 단어로 이루어져 있으므로 변수의 수가 매우 많은 고차원 데이터에 해당된다. 이러한 고차원 데이터에서는 계산 효율성과 통계분석의 정확성을 높이기 위해 많은 변수 중 중요한 변수를 선택하기 위한 절차를 거치는 경우가 많다. 텍스트 데이터에서도 많은 단어 중 중요한 단어를 선택하기 위해 여러가지 방법들이 사용되고 있다. 이 연구에서는 단어 선택을 위한 대표적인 필터링 방법인 카이제곱통계량과 정보이득의 공통점과 차이점을 살펴보고 실제 텍스트 데이터에서 이 단어선택 방법들의 성질을 확인해보았다. 카이제곱통계량과 정보이득은 비음성, 볼록성 등의 성질을 공유하지만 불균형 텍스트 데이터에서 카이제곱통계량이 양변수 위주로 단어를 선택하는 반면, 정보이득은 음변수도 상대적으로 많이 선택하는 경향이 있음을 확인하였다.

• 한국컴퓨터정보학회논문지
• /
• 제27권8호
• /
• pp.41-47
• /
• 2022

본 연구는 일목요연하게 정리되지 않은 구어체(대화)에 대한 텍스트 자동 요약 모델을 적용시키는 방법을 제안한다. 본 연구에서는 구어체 데이터인 SAMSum 데이터를 활용하였고, 선행연구에서 문어체 데이터 텍스트 자동 요약 모델 연구에서 제안한 BERTSumExtAbs 모델을 적용하였다. SAMSum 데이터셋은 70% 이상은 두 사람 간 대화, 나머지 약 30%는 세 사람 이상 간 대화로 구성되어 있다. 본 논문에서는 텍스트 자동 요약 모델을 구어체 데이터에 적용하여, ROUGE Score R-1 부문에서 42.43 이상의 결과를 도출해내었다. 또한, 텍스트 요약 모델로 기존에 제안된 모델인 BERTSum 모델을 fine-tuning하여, 45.81의 높은 점수를 도출했다. 본 연구를 통하여 구어체 데이터에 대한 텍스트 생성 요약의 성능을 입증하였으며, 앞으로 사람의 자연어를 있는 그대로 컴퓨터가 이해하여 다양한 task를 해결하는 데 기초 자료로 활용되길 바란다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.5-8
• /
• 2001

The computer programs based upon graphic user interface(GUI) became commonplace with the advance of computer technology. Nevertheless, programs for the visually-handicapped have still remained at the level of TTS(text to speech) programs and this prevents many visually-handicapped from enjoying the pleasure and convenience of the information age. This paper is, paying attention to the importance of character recognition in images, about the configuration of the system that converts text in the image selected by a user to the speech by extracting the character part, and carrying out character recognition.

• 한국도서관정보학회지
• /
• 제30권3호
• /
• pp.27-51
• /
• 1999

This paper deals with the methods as well as the problems associated with automatic extraction indexing and assignment indexing, expert systems for indexing, and major approaches currently used to index the Internet resources. It also briefly reviews basic methods for establishing hypertext/hypermedia links automatically. The methods used in much of text processing today are not particularly new. Most of the them were used, perhaps in a more rudimentary form, 30 or more years ago by Luhn and many other investigators. Better results can be achieved today because much greater bodies of electronic text are now avaliable and the power of present-day computers allows the processing of such text with reasonable efficiency.

• Journal of the Korean Data and Information Science Society
• /
• 제14권4호
• /
• pp.817-824
• /
• 2003

We employed a hierarchical document classification method to classify a massive collection of electronic medical records(EMR) written in both Korean and English. Our experimental system has been learned from 5,000 records of EMR text data and predicted a newly given set of EMR text data over 68% correctly. We expect the accuracy rate can be improved greatly provided a dictionary of medical terms or a suitable medical thesaurus. The classification system might play a key role in some clinical decision support systems and various interpretation systems for clinical data.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제53권8호
• /
• pp.606-610
• /
• 2004

A noble caption text detection and recognition algorithm using the temporal nature of video is proposed in this paper. A text registration technique is used to locate the temporal and spatial positions of captions in video from the accumulated frame difference information. Experimental results show that the proposed method is effective and robust. Also, a high processing speed is achieved since no time consuming operation is included.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권7호
• /
• pp.1807-1822
• /
• 2023

Innovation and rapid increased functionality in user friendly smartphones has encouraged shutterbugs to have picturesque image macros while in work environment or during travel. Formal signboards are placed with marketing objectives and are enriched with text for attracting people. Extracting and recognition of the text from natural images is an emerging research issue and needs consideration. When compared to conventional optical character recognition (OCR), the complex background, implicit noise, lighting, and orientation of these scenic text photos make this problem more difficult. Arabic language text scene extraction and recognition adds a number of complications and difficulties. The method described in this paper uses a two-phase methodology to extract Arabic text and word boundaries awareness from scenic images with varying text orientations. The first stage uses a convolution autoencoder, and the second uses Arabic Character Segmentation (ACS), which is followed by traditional two-layer neural networks for recognition. This study presents the way that how can an Arabic training and synthetic dataset be created for exemplify the superimposed text in different scene images. For this purpose a dataset of size 10K of cropped images has been created in the detection phase wherein Arabic text was found and 127k Arabic character dataset for the recognition phase. The phase-1 labels were generated from an Arabic corpus of quotes and sentences, which consists of 15kquotes and sentences. This study ensures that Arabic Word Awareness Region Detection (AWARD) approach with high flexibility in identifying complex Arabic text scene images, such as texts that are arbitrarily oriented, curved, or deformed, is used to detect these texts. Our research after experimentations shows that the system has a 91.8% word segmentation accuracy and a 94.2% character recognition accuracy. We believe in the future that the researchers will excel in the field of image processing while treating text images to improve or reduce noise by processing scene images in any language by enhancing the functionality of VGG-16 based model using Neural Networks.

• 대한교통학회지
• /
• 제27권5호
• /
• pp.63-74
• /
• 2009

본 연구에서는 실시간 주행환경에서 전방의 위험상황에 대해 운전자가 효과적으로 반응 할 수 있는 교통안전 경고정보 컨텐츠 도출을 위한 연구를 수행하였다. 차내 단말기를 통해 제공되는 경고정보 유형에 따른 운전자 반응특성을 분석하였다. 운전자 반응특성자료를 수집하기 위하여 DGPS를 장착한 실험차량을 이용하여 개별차량 주행궤적을 수집하여 속도 및 가감속도를 산출하였다. 차내 경고정보 유형은 사고위험성과 직접적인 연관이 있는 긴급영향권과 긴급영향권에 진입하는 차량에게 위험발생의 사전정보를 제공하기 위한 일반영향권으로 분류하였다. 컨텐츠는 시각적 정보인 Text, 이미지와 청각적 정보인 경고음, 그리고 음성 메시지를 고려하여 다양한 시나리오를 제작하여 현장실험에 활용하였다. 분석 결과 긴급영향권에서는 음성+이미지+text와 경고음+text로 구성된 경고정보를 제공하는 것이 가장 효과적인 것으로 분석되었으며, 일반영향권에서는 경고음+음성+이미지+text로 구성된 경고정보를 제공하는 것이 가장 효과적인 것으로 나타났다. 본 연구 결과는 향후 운전자의 반응특성을 고려한 보다 효과적인 실시간 경고정보를 설계하기 위한 유용한 정보로 활용 될 것으로 기대된다.