• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권10호
• /
• pp.4072-4091
• /
• 2015

A separation of text and non-text elements plays an important role in document layout analysis. A number of approaches have been proposed but the quality of separation result is still limited due to the complex of the document layout. In this paper, we present an efficient method for the classification of text and non-text components in document image. It is the combination of whitespace analysis with multi-layer homogeneous regions which called recursive filter. Firstly, the input binary document is analyzed by connected components analysis and whitespace extraction. Secondly, a heuristic filter is applied to identify non-text components. After that, using statistical method, we implement the recursive filter on multi-layer homogeneous regions to identify all text and non-text elements of the binary image. Finally, all regions will be reshaped and remove noise to get the text document and non-text document. Experimental results on the ICDAR2009 page segmentation competition dataset and other datasets prove the effectiveness and superiority of proposed method.

• 대한음성학회지:말소리
• /
• 제53호
• /
• pp.75-92
• /
• 2005

Prosody Generation Based on C-ToBI Representation for Text-to-SpeechSeungwon Kim, Yu Zheng, Gary Geunbae Lee, Byeongchang KimProsody modeling is critical in developing text-to-speech (TTS) systems where speech synthesis is used to automatically generate natural speech. In this paper, we present a prosody generation architecture based on Chinese Tone and Break Index (C-ToBI) representation. ToBI is a multi-tier representation system based on linguistic knowledge to transcribe events in an utterance. The TTS system which adopts ToBI as an intermediate representation is known to exhibit higher flexibility, modularity and domain/task portability compared with the direct prosody generation TTS systems. However, the cost of corpus preparation is very expensive for practical-level performance because the ToBI labeled corpus has been manually constructed by many prosody experts and normally requires a large amount of data for accurate statistical prosody modeling. This paper proposes a new method which transcribes the C-ToBI labels automatically in Chinese speech. We model Chinese prosody generation as a classification problem and apply conditional Maximum Entropy (ME) classification to this problem. We empirically verify the usefulness of various natural language and phonology features to make well-integrated features for ME framework.

• 한국IT서비스학회지
• /
• 제15권4호
• /
• pp.1-24
• /
• 2016

With the innovation of information technology, non-face-to-face robo advisor with high accessibility and convenience is spreading. The current robot advisor recommends appropriate investment products after understanding the investment propensity based on the structured data entered directly or indirectly by individuals. However, it is an inconvenient and obtrusive way for financial consumers to inquire or input their own subjective propensity to invest. Hence, this study proposes a way to deduce the propensity to invest in unstructured data that customers voluntarily exposed during consultation or online. Since prediction performance based on unstructured document differs according to the characteristics of text, in this study, classification algorithm optimized for the characteristic of text left by financial consumers is selected by performing prediction performance evaluation of various learning discrimination algorithms and proposed an intelligent method that automatically recommends investment products. User tests were given to MBA students. After showing the recommended investment and list of investment products, satisfaction was asked. Financial consumers' satisfaction was measured by dividing them into investment propensity and recommendation goods. The results suggest that the users high satisfaction with investment products recommended by the method proposed in this paper. The results showed that it can be applies to non-face-to-face robo advisor.

• International Journal of Computer Science & Network Security
• /
• 제21권8호
• /
• pp.87-96
• /
• 2021

The demand for e-learning through video lectures is rapidly increasing due to its diverse advantages over the traditional learning methods. This led to massive volumes of web-based lecture videos. Indexing and retrieval of a lecture video or a lecture video topic has thus proved to be an exceptionally challenging problem. Many techniques listed by literature were either visual or audio based, but not both. Since the effects of both the visual and audio components are equally important for the content-based indexing and retrieval, the current work is focused on both these components. A framework for automatic topic-based indexing and search depending on the innate content of the lecture videos is presented. The text from the slides is extracted using the proposed Merged Bounding Box (MBB) text detector. The audio component text extraction is done using Google Speech Recognition (GSR) technology. This hybrid approach generates the indexing keywords from the merged transcripts of both the video and audio component extractors. The search within the indexed documents is optimized based on the Naïve Bayes (NB) Classification and K-Means Clustering models. This optimized search retrieves results by searching only the relevant document cluster in the predefined categories and not the whole lecture video corpus. The work is carried out on the dataset generated by assigning categories to the lecture video transcripts gathered from e-learning portals. The performance of search is assessed based on the accuracy and time taken. Further the improved accuracy of the proposed indexing technique is compared with the accepted chain indexing technique.

• International Journal of Computer Science & Network Security
• /
• 제22권7호
• /
• pp.65-74
• /
• 2022

In natural language processing, commonly used words such as prepositions are referred to as stop-words; they have no inherent meaning and are therefore ignored in indexing and retrieval tasks. The removal of stop-words from Arabic text has a significant impact in terms of reducing the size of a cor- pus text, which leads to an improvement in the effectiveness and performance of Arabic-language processing systems. This study investigated the effectiveness of applying a stop-word lists elimination with normalization as a preprocessing step. The idea was to merge statistical method with the linguistic method to attain the best efficacy, and comparing the effects of this two-pronged approach in reducing corpus size for Ara- bic natural language processing systems. Three stop-word lists were considered: an Arabic Text Lookup Stop-list, Frequency- based Stop-list using Zipf's law, and Combined Stop-list. An experiment was conducted using a selected file from the Arabic Newswire data set. In the experiment, the size of the cor- pus was compared after removing the words contained in each list. The results showed that the best reduction in size was achieved by using the Combined Stop-list with normalization, with a word count reduction of 452930 and a compression rate of 30%.

• 산업공학
• /
• 제18권1호
• /
• pp.10-21
• /
• 2005

With the increasing proliferation of World Wide Web, electronic mail systems have become very widely used communication tools. Researches on e-mail classification have been very important in that e-mail classification system is a major engine for e-mail response management systems which mine unstructured e-mail messages and automatically categorize them. In this research we compare the performance of Naive Bayesian learning and Centroid-Based Classification using the different data set of an on-line shopping mall and a credit card company. We analyze which method performs better under which conditions. We compared classification accuracy of them which depends on structure and size of train set and increasing numbers of class. The experimental results indicate that Naive Bayesian learning performs better, while Centroid-Based Classification is more robust in terms of classification accuracy.

• 정보처리학회논문지B
• /
• 제15B권5호
• /
• pp.457-464
• /
• 2008

실제 운용 환경에서 자동문서분류시스템의 성공을 위해서 충분하지 못한 학습문서의 문제와 특징 공간들에 대한 사전지식이 없는 상황을 해결하는 것이 관건이다. 이런 맥락에서 많은 자동문서분류 시스템의 구축을 위해 나이브 베이즈 문서분류 알고리즘을 사용한다. 이는 기존 학습된 분류모델과 특징 공간을 점진적으로 갱신함으로써 분류모델을 향상시키는 것이 매우 용이하기 때문이다. 본 논문에서는 특징 가중치를 이용하여 문서분류기의 성능을 향상시키는 기법을 제안한다. 기본 아이디어는 문서분류 모델의 인자로서 특징들의 분포뿐만 아니라 각 특징들의 중요도를 반영하는 것이다. 속성 선택을 미리 수행하여 학습모델을 만드는 것이 아니라, 속성 중요도를 나이브 베이즈 학습 모델에 포함시킴으로써 보다 정확한 모델을 생성할 수 있다. 또한 동적 환경에서 점진적인 특징 가중치 부여를 위해 기존의 특징 갱신 기법을 확장한 알고리즘도 제안한다. 본 논문에서 제안된 기법을 평가하기 위해서 Reuters-21578과 20Newsgroup 문서집합 이용한 실험을 실시하여, 제안된 기법이 전통적인 나이브 베이즈 분류기의 성능을 크게 향상시킴을 증명한다.

• 스마트미디어저널
• /
• 제5권3호
• /
• pp.30-34
• /
• 2016

Nowadays, there are a lot of Korean documents, which often need to be identified in one of printed or handwritten text. Early methods for the identification use structural features, which can be simple and easy to apply to text of a specific font, but its performance depends on the font type and characteristics of the text. Recently, the bag-of-words model has been used for the identification, which can be invariant to changes in font size, distortions or modifications to the text. The method based on bag-of-words model includes three steps: word segmentation using connected component grouping, feature extraction, and finally classification using SVM(Support Vector Machine). In this paper, bag-of-words model based method is proposed using SURF(Speeded Up Robust Feature) for the identification of machine printed and handwritten text in Korean documents. The experiment shows that the proposed method outperforms methods based on structural features.

• 한국ITS학회 논문지
• /
• 제9권5호
• /
• pp.80-86
• /
• 2010

본 논문에서는 차량 내의 멀티미디어 시스템에 장착되는 엔터테인먼트 기능 중의 하나인 음악스토리 자동생성 기술을 소개한다. 음악스토리 비디오 자동생성 기술은 개인이 소지하고 있는 휴대폰을 차량 내의 멀티미디어 시스템과 연결하여, 휴대폰 안에 저장된 음악과 사진의 결합을 통해 음악비디오를 자동으로 생성하는 멀티미디어 요소기술로서, 사용자에게 분위기에 맞게 음악을 들으면서 생성된 음악스토리 비디오를 즐기는 기능을 제공한다. 음악스토리 비디오 자동생성 기술에 대한 성능은 음악분류, 사진분류, 핵심단어 검출 등의 정확도와 생성된 음악스토리 비디오를 시청한 사용자의 MOS 결과를 통해 측정되었다.

• 스마트미디어저널
• /
• 제5권1호
• /
• pp.61-68
• /
• 2016

최근 스마트폰과 태블릿 PC 등의 스마트 기기들의 발전으로 인해 SNS(Social Network Service) 사용자가 증가함에 따라 SNS 정보를 이용한 사용자 감정 분류 방법에 대한 기법들이 활발하게 연구되고 있다. 사용자 감정 분류는 SNS 게시글의 텍스트, 이미지 등을 이용하여 감정을 분류하는 것을 말한다. 본 논문에서는 텍스트에서 대표 형용사를 추출하고 이미지에서 Canny 알고리즘과 삼각함수를 이용해 대표 도형에 대한 값을 추출하여 사용자의 감정을 분류하는 방법을 제안한다. 텍스트에서 추출한 대표 형용사는 텍스트에서 추출한 형용사 중에 빈도수가 가장 높은 형용사로 선정하였으며, 영어 감정어휘 사전인 SentiWordNet을 이용하여 긍정-부정의 수치를 측정했다. 이미지에서 추출되는 도형에서 삼각형, 사각형, 원중에 추출되는 도형을 대표 도형으로 선정했으며, 대표 도형의 종류와 기울기에 따라 쾌-불쾌 수치를 측정하여 사용자의 감정을 분류했다. 최종적으로 Plutchik의 감정 바퀴를 긍정-부정과 쾌-불쾌의 수치를 나타내는 x축과 y축을 갖는 좌표평면으로 재정의하고 대표 형용사와 대표 도형의 값을 재정의한 Plutchik의 감정 바퀴의 좌표 평면에 나타내어 사용자의 감정 분류를 수행한다.