• Journal of Electrical Engineering and information Science
• /
• 제1권2호
• /
• pp.1-8
• /
• 1996

This paper presents a text-driven discourse analysis system, called DPAS. DPAS constructs a discourse structure by weaving together clauses in the text by finding discourse relations between a clause and the clauses in a context. The basic processing model of DPAS is based on the stack based model of discourse analysis suggested by Grosz and Sidner. We extend the model with dynamic programming method to handle various discourse ambiguities effectively and efficiently. We develop the idea of a context space to keep all information of a context. DPAS parses a text by considering all possible discourse relations between a clause and a context. Since different discourse relations may result in different states of a context, DPAS maintains multiple context spaces for an ambiguous text. Since maintaining all interpretations until the whole text is processed requires too much computing resources, DPAS uses the idea of depth-limited search to limit the search space. If there is more than one discourse relation between an input clause and a context, DPAS constructs context spaces one context space for each discourse relation. Then, DPAS applies heuristics to choose the most desirable context space after it processes some more input clauses. Since the basic idea of DPAS is domain independent, although we used descriptive texts to demonstrate DPAS, we believe the idea of DPAS can be extended to understand other styles of texts.

• 지능정보연구
• /
• 제20권2호
• /
• pp.39-58
• /
• 2014

빅데이터의 중요성에 대한 인식이 확산되고, 관련한 기술이 발전됨에 따라, 최근에는 빅데이터의 처리와 분석의 결과를 어떻게 시각화할 것인지가 매우 관심 받는 주제로 부각되고 있다. 이는 분석된 결과를 보다 명확하고 효과적으로 전달하는 데에 있어서 데이터의 시각화가 매우 효과적인 방법이기 때문이다. 시각화는 분석 시스템과 사용자가 소통하기 위한 하나의 그래픽 사용자 인터페이스(GUI)를 담당하는 역할을 한다. 통상적으로 이러한 GUI 부분은 데이터의 처리나 분석의 결과와 독립될 수록 시스템의 개발과 유지보수가 용이하며, MVC(Model-View-Controller)와 같은 디자인 패턴의 적용을 통해 GUI와 데이터 처리 및 관리 부분 간의 결합도를 최소화하는 것이 중요하다. 한편 빅데이터는 크게 정형 데이터와 비정형 데이터로 구분할 수 있는데 정형 데이터는 시각화가 상대적으로 용이한 반면, 비정형 데이터는 시각화를 구현하기가 복잡하고 다양하다. 그럼에도 불구하고 비정형 데이터에 대한 분석과 활용이 점점 더 확산됨에 따라, 기존의 전통적인 정형 데이터를 위한 시각화 도구들의 한계를 벗어나기 위해 각각의 시스템들의 목적에 따라 고유의 방식으로 시각화 시스템이 구축되는 현실에 직면해 있다. 더욱이나 현재 비정형 데이터 분석의 대상 중 대부분을 차지하고 있는 텍스트 데이터의 경우 언어 분석, 텍스트 마이닝, 소셜 네트워크 분석 등 적용 기술이 매우 다양하여 하나의 시스템에 적용된 시각화 기술을 다른 시스템에 적용하는 것이 용이하지 않다. 이는 현재의 텍스트 분석 결과에 대한 정보 모델이 서로 다른 시스템에 적용될 수 있도록 설계되지 못하는 경우가 많기 때문이다. 본 연구에서는 이러한 문제를 해결하기 위하여 다양한 텍스트 데이터 분석 사례와 시각화 사례들의 공통적 구성 요소들을 식별하여 표준화된 정보 모델인 텍스트 데이터 시각화 모델을 제시하고, 이를 통해 시각화의 GUI 부분과 연결할 수 있는 시스템 모델로서의 시각화 프레임워크인 TexVizu를 제안하고자 한다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2002년도 하계학술발표대회 논문집 제21권 1호
• /
• pp.87-90
• /
• 2002

This paper analyzes statistically the relationship between size and balance of text corpus by evaluation of the effect of interview sentences in language model for Korean broadcast news transcription system. Our Korean broadcast news transcription system's ultimate purpose is to recognize not interview speech, but the anchor's and reporter's speech in broadcast news show. But the gathered text corpus for constructing language model consists of interview sentences a portion of the whole, $15\%$ approximately. The characteristic of interview sentence is different from the anchor's and the reporter's in one thing or another. Therefore it disturbs the anchor and reporter oriented language modeling. In this paper, we evaluate the effect of interview sentences in language model for Korean broadcast news transcription system and analyze statistically the relationship between size and balance of text corpus by making an experiment as the same procedure according to varying the size of corpus.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 춘계학술발표대회
• /
• pp.438-439
• /
• 2019

사회적 문제와 인공지능 기술의 발달로 챗봇 서비스에 대한 관심이 점점 증가하고 있으며, 그 결과 TTS(Text to Speech) 및 STT(Speech to Text) 기술을 기반으로 한 보조형 프로그램에 대한 개발이 다양한 모바일 환경에서 진행중이다. 본 논문에서는 문자를 소리로 변환해주는 TTS(Text to Speech) 기술과 소리를 문자로 변환해주는 STT(Speech to Text) 기술을 사용하여 음성지원 챗봇 시스템을 제작하고 이를 안드로이드 기반의 모바일 애플리케이션으로 구현한 '음성지원 챗봇 모바일 애플리케이션'을 제안하고, 이와 관련하여 관련 기술 및 기대효과에 대해 소개한다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2013년도 추계학술발표대회
• /
• pp.768-770
• /
• 2013

We propose a symmetric key based cryptography algorithm to encode and decode the text data with limited length using 3-dimensional magic square matrix. To encode the plain text message, input text will be translated into an index of the number stored in the key matrix. Then, Caesar's shift with pre-defined constant value is fabricated to finalize an encryption algorithm. In decode process, Caesar's shift is applied first, and the generated key matrix is used with 2D magic squares to replace the index numbers in ciphertext to restore an original text.

• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1437-1446
• /
• 2020

For the last few years, smart devices have begun to occupy an essential place in the life of children, by allowing them to access a variety of language activities and books. Various studies are being conducted on using smart devices for education. Our study extracts images and texts from kids' book with smart devices and matches the extracted images and texts to create new images that are not represented in these books. The proposed system will enable the use of smart devices as educational media for children. A deep convolutional generative adversarial network (DCGAN) is used for generating a new image. Three steps are involved in training DCGAN. Firstly, images with 11 titles and 1,164 images on ImageNet are learned. Secondly, Tesseract, an optical character recognition engine, is used to extract images and text from kids' book and classify the text using a morpheme analyzer. Thirdly, the classified word class is matched with the latent vector of the image. The learned DCGAN creates an image associated with the text.

• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권3호
• /
• pp.591-609
• /
• 2024

In this study, preprocessings with all combinations were examined in terms of the effects on decreasing word number, shortening the duration of the process and the classification success in balanced and imbalanced datasets which were unbalanced in different ratios. The decreases in the word number and the processing time provided by preprocessings were interrelated. It was seen that more successful classifications were made with Turkish datasets and English datasets were affected more from the situation of whether the dataset is balanced or not. It was found out that the incorrect classifications, which are in the classes having few documents in highly imbalanced datasets, were made by assigning to the class close to the related class in terms of topic in Turkish datasets and to the class which have many documents in English datasets. In terms of average scores, the highest classification was obtained in Turkish datasets as follows: with not applying lowercase, applying stemming and removing stop words, and in English datasets as follows: with applying lowercase and stemming, removing stop words. Applying stemming was the most important preprocessing method which increases the success in Turkish datasets, whereas removing stop words in English datasets. The maximum scores revealed that feature selection, feature size and classifier are more effective than preprocessing in classification success. It was concluded that preprocessing is necessary for text classification because it shortens the processing time and can achieve high classification success, a preprocessing method does not have the same effect in all languages, and different preprocessing methods are more successful for different languages.

• 디지털콘텐츠학회 논문지
• /
• 제17권4호
• /
• pp.219-226
• /
• 2016

일반 생활 속에서 우리 인간의 눈으로 정보를 인식하고 그 정보를 이용하는 것에는 한계가 없을 만큼 다양하고 방대하다. 그러나 인공지능이 발달한 현재의 기술로도, 인간의 시각 처리 능력에 비하면 턱없이 능력이 부족하다. 그럼에도 불구하고 많은 연구자들은 실생활 속에서 정보를 얻고자 하고 있고, 특히 글자로 된 정보를 인식하는데 많은 노력을 기울이고 있다. 글자를 인식하는 분야에서 일반적인 문서에서 글자를 추출하는 것은 일부 정보처리 분야에서 이용되고 있지만, 실영상에서 문자를 추출하고 인식하는 부분은 아직도 많이 부족하다. 그 이유는 실영상에서는 색깔, 크기, 방향, 공통점 등에서 다양한 특징을 갖고 있기 때문이다. 본 논문에서는 이런 다양한 환경에서 문자 영역을 추출하기 위하여 적응적 에지 강화 기반의 MSER을 적용하여 장면 텍스트 추출을 시도하고, 비교적 좋은 방법임을 실험으로 보인다.

• International Journal of Contents
• /
• 제8권1호
• /
• pp.30-38
• /
• 2012

Text extraction and binarization are the important pre-processing steps for text recognition. The performance of text binarization strongly related to the accuracy of recognition stage. In our proposed method, the first stage based on line detection and shape feature analysis applied to locate the position of a business card and detect the shape from the complex environment. In the second stage, several local regions contained the possible text components are separated based on the projection histogram. In each local region, the pixels grouped into several connected components based on the connected component labeling and projection histogram. Then, classify each connect component into text region and reject the non-text region based on the feature information analysis such as size of connected component and stroke width estimation.