• International Journal of Computer Science & Network Security
• /
• v.22 no.1
• /
• pp.283-287
• /
• 2022

Machine Learning is the most popular method used in data science. Growth of data is not only numeric data but also text data. Most of the algorithm of supervised and unsupervised machine learning algorithms use numeric data. Now it is required to convert text data into numeric. There are many techniques for this conversion. Researcher confuses which technique is best in what situation. Here in proposed work BOW (Bag-of-Words) and TF-IDF (Term-Frequency-Inverse-Document-Frequency) has been studied based on different features to determine best method. After experimental results on text data, TF-IDF and BOW both provide better performance at range from 100 to 150 number of features.

• International Journal of Computer Science & Network Security
• /
• v.24 no.1
• /
• pp.205-208
• /
• 2024

Artificial Intelligence (AI) technology has evolved rapidly in recent years and is used in everything from banking to email management to surgery, but without the help of the visible, most of the fun features of the Internet include visual impairment. It benefits people with disabilities. The main purpose of this study is to find ways to help people with visual impairments using AI technology. A visually impaired request is made for the visually impaired. For example, when a message arrives that the program will notify you by voice (reads the sender's name, read the message, and replies to it if necessary), this is a special program installed on your mobile phone. This program uses a customized algorithm developed in Python to convert written text to voice, read text, and convert voice to written text on a message when a visually impaired person wants to respond. Then it sends the response in the form of a text message. Therefore, the research should lead to programs for people with visual impairments. This program makes mobile phones easier and more comfortable to use and makes the daily life easier for visual impairments.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.5
• /
• pp.700-707
• /
• 2017

Recently, the military need more various education and training because of the increasing necessity of various operation. But the education and training of the military has the various difficulties such as the limitations of time, space and finance etc. In order to overcome the difficulties, the military use Defense Modeling and Simulation(DM&S). Although the participants in training has the empirical knowledge from education and training based on the simulation, the empirical knowledge is not shared because of particular characteristics of military such as security and the change of official. This situation obstructs the improving effectiveness of education and training. The purpose of this research is the systematizing and analysing the empirical knowledge using text mining and network analysis to assist the sharing of empirical knowledge. For analysing texts or documents as the empirical knowledge, we select the text mining and network analysis. We expect our research will improve the effectiveness of education and training based on simulation of DM&S.

• Korea Trade Review
• /
• v.45 no.6
• /
• pp.23-43
• /
• 2020

This study aims at analyzing the interactions between two of the most significant trade stakeholders in Korea, the Trade Ministry and the Legislature, using text network analysis. Tackling seven Action and Plan Reports for Requests from Parliamentary Inspection released by the National Assembly, this paper conducts a topic modelling analysis, particularly focusing on the reports for the three trade-related institutes: the MOTIE headquarter, Korea Trade Insurance Corporation, Korea Trade and Investment Promotion Agency. According to the analysis, such traditional topics of the MOTIE as enterprise, industry, business, management, development were frequently appeared in the reports. Trade-related topics including export, trade, commerce, investment, overseas, domestic, dispute, cooperation, efficiency, negotiation, service, promotion were repeatedly shown. Lastly, a case study on 2019 Parliamentary Inspection Report showed specific trade-related topics and relevant contents that raised issues in that year. This analysis implies that the text data driven from the Parliamentary Inspection Reports between the MOTIE and the National Assembly, can be established as so called 'trade policy information system' which are valuable not only for the two but also the rest of the trade stakeholders in Korea.

• Journal of the Korean Wood Science and Technology
• /
• v.51 no.6
• /
• pp.458-469
• /
• 2023

Text mining techniques provide valuable insights into research information across various fields. In this study, text mining was used to identify research trends in wood science from 2012 to 2022, with a focus on representative journals published in Korea and Japan. Abstracts from Journal of the Korean Wood Science and Technology (JKWST, 785 articles) and Journal of Wood Science (JWS, 812 articles) obtained from the SCOPUS database were analyzed in terms of the word frequency (specifically, term frequency-inverse document frequency) and co-occurrence network analysis. Both journals showed a significant occurrence of words related to the physical and mechanical properties of wood. Furthermore, words related to wood species native to each country and their respective timber industries frequently appeared in both journals. CLT was a common keyword in engineering wood materials in Korea and Japan. In addition, the keywords "MDF," "MUF," and "GFRP" were ranked in the top 50 in Korea. Research on wood anatomy was inferred to be more active in Japan than in Korea. Co-occurrence network analysis showed that words related to the physical and structural characteristics of wood were organically related to wood materials.

• Proceedings of the Zoological Society Korea Conference
• /
• 1999.10b
• /
• pp.37-37
• /
• 1999

No Abstract, See Full Text

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1759-1772
• /
• 2023

Chinese named entity recognition (NER) is a challenging work that seeks to find, recognize and classify various types of information elements in unstructured text. Due to the Chinese text has no natural boundary like the spaces in the English text, Chinese named entity identification is much more difficult. At present, most deep learning based NER models are developed using a bidirectional long short-term memory network (BiLSTM), yet the performance still has some space to improve. To further improve their performance in Chinese NER tasks, we propose a new NER model, IDCNN-BiLSTM-Highway, which is a combination of the BiLSTM, the iterated dilated convolutional neural network (IDCNN) and the highway network. In our model, IDCNN is used to achieve multiscale context aggregation from a long sequence of words. Highway network is used to effectively connect different layers of networks, allowing information to pass through network layers smoothly without attenuation. Finally, the global optimum tag result is obtained by introducing conditional random field (CRF). The experimental results show that compared with other popular deep learning-based NER models, our model shows superior performance on two Chinese NER data sets: Resume and Yidu-S4k, The F1-scores are 94.98 and 77.59, respectively.

• Journal of Korea Society of Industrial Information Systems
• /
• v.21 no.3
• /
• pp.35-46
• /
• 2016

Social network services (SNS) that help to build relationship network and share a particular interest or activity freely according to their interests by posting comments, photos, videos,${\ldots}$ on online communities such as blogs have adopted and developed widely as a social phenomenon. Several researches have been done to explore the pattern and valuable information in social networks data via text mining such as opinion mining and semantic analysis. For improving the efficiency of text mining, keyword-based approach have been applied but most of researchers argued the limitations of the rules of Korean orthography. This research aims to construct a database of non-standard Korean words which are difficulty in data mining such abbreviations, slangs, strange expressions, emoticons in order to improve the limitations in keyword-based text mining techniques. Based on the study of subjective opinions about specific topics on blogs, this research extracted non-standard words that were found useful in text mining process.

• Journal of Digital Convergence
• /
• v.19 no.3
• /
• pp.401-414
• /
• 2021

The purpose of this study was to examine the associations of Korean older adults' use of Social Network Service (SNS) and text messages with frequency of contact with 1) non-coresident adult children, 2) siblings and relatives, or 3) friends, neighbors, and acquaintances. Data were drawn from the 2017 Survey of Living Conditions and Welfare Needs of Korean Older Persons 65+ (N=8,392), and older adults were categorized into 4 groups depending on their familiarity with use of SNS and text messages. Ordinary Least Squares regression models were estimated for analyses. Results revealed that older users of both types of communication media reported frequent exchanges of calls, text messages, etc. with both family and friends. However, using SNS and text messages was consistently related to more face-to-face contact with non-family members. To conclude, older adults' familiarity with communication media could be key to exchanges of emotional and instrumental support with informal social network members and quality of life in the community. Overall, our results highlight the importance of information communication education targeting older adults for continued involvement with their informal social network members.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.1
• /
• pp.29-34
• /
• 2022

We propose a novel network architecture and build dataset for recognizing clickable objects on mobile device screens. The data was collected based on clickable objects on the mobile device screen that have numerous resolution, and a total of 24,937 annotation data were subdivided into seven categories: text, edit text, image, button, region, status bar, and navigation bar. We use the Deconvolution Single Shot Detector as a baseline, the backbone network with Squeeze-and-Excitation blocks, the Single Shot Detector layer structure to derive inference results and the Feature pyramid networks structure. Also we efficiently extract features by changing the input resolution of the existing 1:1 ratio of the network to a 1:2 ratio similar to the mobile device screen. As a result of experimenting with the dataset we have built, the mean average precision was improved by up to 101% compared to baseline.