• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.479-486
• /
• 2005

The ontology was proposed to construct the logical basis of semantic web. Ontology represents domain knowledge in the formal form and it enables that machine understand domain knowledge and provide appropriate intelligent service for user request. However, the construction and the maintenance of ontology requires large amount of cost and human efforts. This paper proposes an automatic ontology construction method for defining relation between concepts in the documents. The Proposed method works as following steps. First we find concept pairs which compose association rule based on the concepts in domain specific documents. Next, we find pattern that describes the relation between concepts by clustering the context between two concepts composing association rule. Last, find generalized pattern name by clustering the clustered patterns. To verify the proposed method, we extract relation between concepts and evaluate the result using documents set provide by TREC(Text Retrieval Conference). The result shows that proposed method cant provide useful information that describes relation between concepts.

• Journal of the Korea Convergence Society
• /
• v.12 no.4
• /
• pp.31-42
• /
• 2021

The purpose of this study was to explore research topics and examine the trend in COVID19 related research papers. We identified eight topics using latent Dirichlet allocation and found acceptable validity in comparison with the structural topic model. The subtopics have been extracted using k-means clustering and plotted in PCA space. Additionally, we discovered the topics bearing negative tones and warning signs by sentiment analysis. The results flagged up the issues of the topics, Biomedical Related, International Dynamics and Psychological Impact. The findings could serve as a guideline for researchers who explore new research directions and policymakers who need to make decisions about which research projects to support.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.3
• /
• pp.31-39
• /
• 2008

Until now, there are many technologies to improve studying ability using e-learning system. In most of e-learning system, learners are studying through the lecture materials and studying problems. The studying ability and intention, however, can be improved through the shared materials and discussion. In this case, learning materials are shared by the learners' discussion and shared materials through the board Internet and MSN. Such data was not classified by learners; it was not easy for the learners to search related valuable information. Therefore, it was not helping to learning. The technologies of most text mining extract summary data from the collection of document or classify into similar document from the complex document. In this paper, we implemented e-learning system for learners to improve learning abilities and especially, applied text mining technology to classify learning material for helping learners.

• Journal of Internet Computing and Services
• /
• v.20 no.1
• /
• pp.1-10
• /
• 2019

Since big-data text mining extracts many features and data, clustering and classification can result in high computational complexity and low reliability of the analysis results. In particular, a term document matrix obtained through text mining represents term-document features, but produces a sparse matrix. We designed an advanced genetic algorithm (GA) to extract features in text mining for detection model. Term frequency inverse document frequency (TF-IDF) is used to reflect the document-term relationships in feature extraction. Through a repetitive process, a predetermined number of features are selected. And, we used the sparsity score to improve the performance of detection model. If a spam mail data set has the high sparsity, detection model have low performance and is difficult to search the optimization detection model. In addition, we find a low sparsity model that have also high TF-IDF score by using s(F) where the numerator in fitness function. We also verified its performance by applying the proposed algorithm to text classification. As a result, we have found that our algorithm shows higher performance (speed and accuracy) in attack mail classification.

• ETRI Journal
• /
• v.23 no.4
• /
• pp.168-176
• /
• 2001

This paper discusses two important issues of corpus-based synthesis: synthesis unit generation based on phrase break strength information and pruning redundant synthesis unit instances. First, the new sentence set for recording was designed to make an efficient synthesis database, reflecting the characteristics of the Korean language. To obtain prosodic context sensitive units, we graded major prosodic phrases into 5 distinctive levels according to pause length and then discriminated intra-word triphones using the levels. Using the synthesis unit with phrase break strength information, synthetic speech was generated and evaluated subjectively. Second, a new pruning method based on weighted vector quantization (WVQ) was proposed to eliminate redundant synthesis unit instances from the synthesis database. WVQ takes the relative importance of each instance into account when clustering similar instances using vector quantization (VQ) technique. The proposed method was compared with two conventional pruning methods through objective and subjective evaluations of synthetic speech quality: one to simply limit the maximum number of instances, and the other based on normal VQ-based clustering. For the same reduction rate of instance number, the proposed method showed the best performance. The synthetic speech with reduction rate 45% had almost no perceptible degradation as compared to the synthetic speech without instance reduction.

• Journal of Biomedical Engineering Research
• /
• v.9 no.1
• /
• pp.93-100
• /
• 1988

The Auto Regressive Parameter Estimation and Pattern Classification of EKG Signal for Automatic Diagnosis. This paper presents the results from pattern discriminant analysis of an AR (auto regressive) model parameter group, which represents the HRV (heart rate variability) that is being considered as time series data. HRV data was extracted using the correct R-point of the EKG wave that was A/D converted from the I/O port both by hardware and software functions. Data number (N) and optimal (P), which were used for analysis, were determined by using Burg's maximum entropy method and Akaike's Information Criteria test. The representative values were extracted from the distribution of the results. In turn, these values were used as the index for determining the range o( pattern discriminant analysis. By carrying out pattern discriminant analysis, the performance of clustering was checked, creating the text pattern, where the clustering was optimum. The analysis results showed first that the HRV data were considered sufficient to ensure the stationarity of the data; next, that the patern discrimimant analysis was able to discriminate even though the optimal order of each syndrome was dissimilar.

• Korean Journal of Acupuncture
• /
• v.36 no.4
• /
• pp.292-299
• /
• 2019

Objectives : DongUiBoGam is one of the representative medical literatures in Korea. We used text mining methods and analyzed the characteristics of the indications of each meridian in the second chapter of DongUiBoGam, WaeHyeong, which addresses external body elements. We also visualized the relationships between the meridians and the disease sites. Methods : Using the term frequency-inverse document frequency (TF-IDF) method, we quantified values regarding the indications of each meridian according to the frequency of the occurrences of 14 meridians and 14 disease sites. The spatial patterns of the indications of each meridian were visualized on a human body template according to the TF-IDF values. Using hierarchical clustering methods, twelve meridians were clustered into four groups based on the TF-IDF distributions of each meridian. Results : TF-IDF values of each meridian showed different constellation patterns at different disease sites. The spatial patterns of the indications of each meridian were similar to the route of the corresponding meridian. Conclusions : The present study identified spatial patterns between meridians and disease sites. These findings suggest that the constellations of the indications of meridians are primarily associated with the lines of the meridian system. We strongly believe that these findings will further the current understanding of indications of acupoints and meridians.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06c
• /
• pp.62-65
• /
• 2010

Image annotations allow users to access a large image database with textual queries. But since the surrounding text of Web images is generally noisy. an efficient image annotation and retrieval system is highly desired. which requires effective image search techniques. Data mining techniques can be adopted to de-noise and figure out salient terms or phrases from the search results. Clustering algorithms make it possible to represent visual features of images with finite symbols. Annotationbased image search engines can obtains thousands of images for a given query; but their results also consist of visually noise. In this paper. we present a new algorithm Double-Circles that allows a user to remove noise results and characterize more precise representative annotations. We demonstrate our approach on images collected from Flickr image search. Experiments conducted on real Web images show the effectiveness and efficiency of the proposed model.

• Journal of Advanced Navigation Technology
• /
• v.22 no.6
• /
• pp.670-674
• /
• 2018

Social network service (SNS) related data including comments/text, images, videos, blogs, and user experiences contain a wealth of information which can be used to build recommendation systems for various clients' and provide insightful data/results to business analysts. Multimedia data, especially visual data like image and videos are the richest source of SNS data which can reflect particular region, and cultures values/interests, form a gigantic portion of the overall data. Mining such huge amounts of data for extracting actionable intelligence require efficient and smart data analysis methods. The purpose of this paper is to focus on this particular modality for devising ways to model, index, and retrieve data as and when desired.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.420-438
• /
• 2004

Artificially or naturally contained texts in the natural images have significant and detailed information about the scenes. If we develop a method that can extract and recognize those texts in real-time, the method can be applied to many important applications. In this paper, we suggest a new method that extracts the text areas in the natural images using the low-level image features of color continuity. gray-level variation and color valiance and that verifies the extracted candidate regions by using the high-level text feature such as stroke. And the two level features are combined hierarchically. The color continuity is used since most of the characters in the same text lesion have the same color, and the gray-level variation is used since the text strokes are distinctive in their gray-values to the background. Also, the color variance is used since the text strokes are distinctive in their gray-values to the background, and this value is more sensitive than the gray-level variations. The text level stroke features are extracted using a multi-resolution wavelet transforms on the local image areas and the feature vectors are input to a SVM(Support Vector Machine) classifier for the verification. We have tested the proposed method using various kinds of the natural images and have confirmed that the extraction rates are very high even in complex background images.