• Journal of KIISE:Software and Applications
• /
• v.29 no.6
• /
• pp.417-424
• /
• 2002

Automatic text categorization is a problem of assigning predefined categories to free text documents. In order to classify text documents, we have to extract good features from them. In previous researches, a text document is commonly represented by the frequency of each feature. But there is a difference between important and unimportant sentences in a text document. It has an effect on the importance of features in a text document. In this paper, we measure the importance of sentences in a text document using text summarizing techniques. A text document is represented by features with different weights according to the importance of each sentence. To verify the new method, we constructed Korean news group data set and experiment our method using it. We found that our new method gale a significant improvement over a basis system for our data sets.

• Journal of Information Technology Applications and Management
• /
• v.12 no.2
• /
• pp.15-27
• /
• 2005

Text categorization systems generally use single words (unigrams) as features. A deceptively simple algorithm for improving text categorization is investigated here, an idea previously shown not to work. It is to identify useful word pairs (bigrams) made up of adjacent unigrams. The bigrams it found, while small in numbers, can substantially raise the quality of feature sets. The algorithm was tested on two pre-classified datasets, Reuters-21578 for English and Korea-web for Korean. The results show that the algorithm was successful in extracting high quality bigrams and increased the quality of overall features. To find out the role of bigrams, we trained the Na$\"{i}$ve Bayes classifiers using both unigrams and bigrams as features. The results show that recall values were higher than those of unigrams alone. Break-even points and F1 values improved in most documents, especially when documents were classified along the large classes. In Reuters-21578 break-even points increased by 2.1%, with the highest at 18.8%, and F1 improved by 1.5%, with the highest at 3.2%. In Korea-web break-even points increased by 1.0%, with the highest at 4.5%, and F1 improved by 0.4%, with the highest at 4.2%. We can conclude that text classification using unigrams and bigrams together is more efficient than using only unigrams.

• Proceedings of the KSPS conference
• /
• 1996.10a
• /
• pp.473-475
• /
• 1996

The prosodic features of Spoken Chinese play the important roll of the naturalness, a list of prosodic labeling symbols represents all the prosodic features is given in this paper, and a paragraph of ' Prosodic Labeling Text '(PLT) is also attached for example.

• Smart Media Journal
• /
• v.5 no.3
• /
• 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.

• Journal of Information Management
• /
• v.33 no.2
• /
• pp.19-32
• /
• 2002

With the growth of the digital library and the use of Internet, the amount of online text information has increased rapidly. The need for efficient data management and retrieval techniques has also become greater. An automatic text categorization system assigns text documents to predefined categories. The system allows to reduce the manual labor for text categorization. In order to classify text documents, the good features from the documents should be selected and the documents are indexed with the features. In this paper, each steps of text categorization and several techniques used in each step are introduced.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.29 no.7 s.144
• /
• pp.926-937
• /
• 2005

The analysis category of Art to Wear was text analyzed from the research material of 100 projects put together by fashion specialist. The conclusion of Art to Wear was comprehended the general features of it were compared and analyzed from a semiotics context. According to this analysis, the formative features of modern Art to Wear is categorized into three different dimensions from a semiotics light. The formative features of modem Art to Wear in the light of syntactic dimension was divided as an open constructed shape of Space Extension, non-typical Deformation, Geometrical Plasticity. The formative features of modem Art to Wear in the light of semantic dimension express symbolic meaning through metaphorical sign. These sign reflect the body image of the life and death and its objective of Abjection, Hybrid of discultural appearance and the image of Hyper-reality, which are features used to comprehend the inner meaning. The formative features of modem Art to Wear in the light of pragmatic dimension divided the artist emotion and meaning system delivered by Emotive Image, the Phatic Image that arouse inner signification and the Poetic Image which contain artistic and aesthetic meaning within it.

• ETRI Journal
• /
• v.43 no.1
• /
• pp.95-108
• /
• 2021

Application (app) ratings are feedback provided voluntarily by users and serve as important evaluation criteria for apps. However, these ratings can often be biased owing to insufficient or missing votes. Additionally, significant differences have been observed between numeric ratings and user reviews. This study aims to predict the numeric ratings of Google apps using machine learning classifiers. It exploits numeric app ratings provided by users as training data and returns authentic mobile app ratings by analyzing user reviews. An ensemble learning model is proposed for this purpose that considers term frequency/inverse document frequency (TF/IDF) features. Three TF/IDF features, including unigrams, bigrams, and trigrams, were used. The dataset was scraped from the Google Play store, extracting data from 14 different app categories. Biased and unbiased user ratings were discriminated using TextBlob analysis to formulate the ground truth, from which the classifier prediction accuracy was then evaluated. The results demonstrate the high potential for machine learning-based classifiers to predict authentic numeric ratings based on actual user reviews.

• Journal of Korea Multimedia Society
• /
• v.20 no.6
• /
• pp.882-892
• /
• 2017

Locally Likely Arrangement Hashing (LLAH) is a method that describes image features based on the geometry between their neighbors. Thus, it has been preferred to implement augmented reality on poorly-textured objects such as text documents. However, LLAH strongly requires that image features be detected with high repeatability and located at a distance from one another. To fulfill the requirement for text document, this paper proposes a method that facilitates the word detection in infrared (IR) range by adjusting the IR-reflectance of words. Specifically, the words are printed out with two different black inks: one is using the K(carbon black) ink only, the other is mixing the C(cyan), M(magenta), Y(yellow) inks. Since only the words printed out with the K ink is visible in IR range, a part of words are selected in advance to be used as features and printed out the K ink. The selected words can be robustly detected with high repeatability in IR range and this enables to implement augmented reality on text documents with high fidelity. The validity of the proposed method was verified through experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3211-3229
• /
• 2022

To solve the problems existing in the process of Weibo disaster rumor recognition, such as lack of corpus, poor text standardization, difficult to learn semantic information, and simple semantic features of disaster rumor text, this paper takes Sina Weibo as the data source, constructs a dataset for Weibo disaster rumor recognition, and proposes a deep learning model BERT_AT_Stacked LSTM for Weibo disaster rumor recognition. First, add adversarial disturbance to the embedding vector of each word to generate adversarial samples to enhance the features of rumor text, and carry out adversarial training to solve the problem that the text features of disaster rumors are relatively single. Second, the BERT part obtains the word-level semantic information of each Weibo text and generates a hidden vector containing sentence-level feature information. Finally, the hidden complex semantic information of poorly-regulated Weibo texts is learned using a Stacked Long Short-Term Memory (Stacked LSTM) structure. The experimental results show that, compared with other comparative models, the model in this paper has more advantages in recognizing disaster rumors on Weibo, with an F1_Socre of 97.48%, and has been tested on an open general domain dataset, with an F1_Score of 94.59%, indicating that the model has better generalization.

• Journal of The Korean Association For Science Education
• /
• v.41 no.2
• /
• pp.71-82
• /
• 2021

Learning efficiency can be maximized by careful matching of text features to expected reader features (i.e., linguistic and cognitive abilities, and background knowledge). The present study aims to explore whether this systematic principle is reflected in the development of science textbooks. The current study examined science textbook texts on 20 measures provided by Auto-Kohesion, a Korean language analysis tool. In addition to surface-level features (basic counts, word-related measures, syntactic complexity measures) which have been commonly used in previous text analysis studies, the present study included cohesion-related features as well (noun overlap ratios, connectives, pronouns). The main findings demonstrate that the surface measures (e.g., word and sentence length, word frequency) overall increased in complexity with grade levels, whereas the majority of the other measures, particularly cohesion-related measures, did not systematically vary across grade levels. The current results suggest that students of lower grades are expected to experience learning difficulties and lowered motivation due to the challenging texts. Textbooks are also not likely to be suitable for students of higher grades to develop the ability to process difficulty level texts required for higher education. The current study suggests that various text-related features including cohesion-related measures need to be carefully considered in the process of textbook development.