• Journal of Korea Multimedia Society
• /
• v.25 no.11
• /
• pp.1557-1563
• /
• 2022

Currently Spatial Data is collected and processed in various methods, and its usability is very high. However, the existing Spatial Data analysis Software usually requires professional knowledge in the collection, refinement, and application of spatial Date, making it difficult to access and apply it. Therefore, this study established a new WebGIS platform with improved accessibility and usability to solve these problems. This platform supports various services such as master map sharing, spatial data generation, automatic coordinate system conversion, WMS issuance, grid generation, and grid analysis. These services increase operational convenience, such as simplifying repetitive tasks and automatically expressing text files. While it is believed that non-experts can easily and conveniently because of them to simplify and express the results. In addition, it is judged to have high accuracy and reliability compared to the analysis results using the existing Open Source-based GIS software.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.224-230
• /
• 2021

YouTube is an OTT service that leads the home economy, which has emerged from the 2020 Corona Pandemic. With the growth of OTT-based individual media, creators are required to establish attractive storytelling strategies that can be preferred by viewers and elected for YouTube recommendation algorithms. In this study, we conducted a study on modeling that proposes a content storyline for creators. As the ability for Creators to create content that viewers prefer, we have presented the data literacy ability to find patterns in complex and massive data. We also studied the importance of compelling storytelling configurations that viewers prefer and can be selected for YouTube recommendation algorithms. This study is of great significance in that it deviated from the viewer-oriented recommendation system method and proposed a story suggestion model for individual creaters. As a result of incorporating this story proposal model into the production of the YouTube channel Tiger Love video, it showed a certain effectiveness. This story suggestion model is a machine learning text-based story suggestion system, excluding the application of photography or video.

• Journal of Industrial Convergence
• /
• v.15 no.1
• /
• pp.33-42
• /
• 2017

The purpose of this study is to identify the current situation of bicycle rental center in Seoul through big data analysis and to find ways to improve it. For this purpose, we analyzed the open data set provided by the Seoul Metropolitan Government and the typical data which is the citizen opinion of the customer center of the Seoul City bicycle. As the result, it was found that it is better to install a bicycle rental shop in Gangdong-gu, Seoul.

• International Journal of Computer Science & Network Security
• /
• v.21 no.11
• /
• pp.23-30
• /
• 2021

Sentiment Analysis has become very important field of research because posting of reviews is becoming a trend. Supervised, unsupervised and semi supervised machine learning methods done lot of work to mine this data. Feature engineering is complex and technical part of machine learning. Deep learning is a new trend, where this laborious work can be done automatically. Many researchers have done many works on Deep learning Convolutional Neural Network (CNN) and Long Shor Term Memory (LSTM) Neural Network. These requires high processing speed and memory. Here author suggested two models simple & bidirectional deep leaning, which can work on text data with normal processing speed. At end both models are compared and found bidirectional model is best, because simple model achieve 50% accuracy and bidirectional deep learning model achieve 99% accuracy on trained data while 78% accuracy on test data. But this is based on 10-epochs and 40-batch size. This accuracy can also be increased by making different attempts on epochs and batch size.

• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.424-427
• /
• 2022

본 연구는 도메인 특성이 강한 HR 평가문장을 BERT PLM 모델을통해 4 가지 class 로 구분하는 문제를 다룬다. 다양한 PLM 모델 적용과 training data 수에 따른 모델 성능 비교를 통해 특정 도메인에 언어모델을 적용하기 위해서 필요한 기준을 확인하였다. 또한 Unlabeled 된 HR 분야 corpus 를 활용하여 BERT 모델을 post-training 한 HR-BERT 가 PLM 분석모델 정확도 향상에 미치는 결과를 탐구한다. 위와 같은 연구를 통해 HR 이 가지고 있는 가장 큰 text data 에 대한 활용 기반을 마련하고, 특수한 도메인 분야에 PLM 을 적용하기 위한 가이드를 제시하고자 한다

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.964-966
• /
• 2022

본 연구는 유료 OCR 서비스를 이용하여 야외 촬영 이미지의 텍스트를 검출하는 프로젝트에서 야외 촬영 텍스트를 학습시킨 Yolov4 모델을 통한 전처리 작업을 제안한다. 텍스트 감지를 통한 이미지 텍스트 전처리 진행은 불필요한 OCR 실행을 줄여 리소스를 절약하고 유료 서비스의 경우 비용 절감 효과까지 도모할 수 있다는 장점이 있다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.2
• /
• pp.421-431
• /
• 1999

NOD systems contain the data of multiple types such as text, image and video, and the size of NOD data depend on their data types. Therefore, in this paper, we propose a Data Placement Method based on Data Types(DPMDT), in which the data placement method depends on their type. Then, we analyze the performance of DPMDT with that of a Time Based Storage Management(TBSM) in which the data placement method depends on their created date, and that of Rate Based Storage Management(RBSM) in which the data placement method depends on their created date and accessed rate. In case of long playback of video news and a few disks(one disk), our results show that the performance of DPMDT is less efficient than that of TBSM and RBSM methods, however, in case of over 2 disks, the performance of DPMDT is more efficient than that of TBSM and RBSM methods.

• International Journal of Knowledge Content Development & Technology
• /
• v.9 no.4
• /
• pp.31-45
• /
• 2019

In this study, a database was established by analyzing the record data research produced in ice hockey sports. The deployed data verification with Ice hockey reference service was demonstrated with ice hockey officials and players. This research utilized the data stored in the KNSU Datanest data repository and developed PDF parsers for batch processing of records. Among the types of records, the game summary, team roster, team statistics, and player statistics files were collected, and tables were extracted from the records. PDF records were converted to text in CSV format which are converted to DataFrame and loaded into the database. Out of the total 22 types of records, 4 types were constructed with OO data parsed as element values. Data verification has found no problems with the quality of the data deployed, showing a high satisfaction with providing 66 factors against the 30 factors provided by the service previously used.

• Korean Journal of Ecology and Environment
• /
• v.48 no.3
• /
• pp.195-202
• /
• 2015

We identified research trends for freshwater exotic species in South Korea using text mining methods in conjunction with bibliometric analysis. We searched scientific and common names of freshwater exotic species as searching keywords including 1 mammal species, 3 amphibian-reptile species, 11 fish species, 2 aquatic plant species. A total of 245 articles including research articles and abstracts of conference proceedings published by 56 academic societies and institutes were collected from scientific article databases. The search keywords used were the common names for the exotic species. The $20^{th}$ century (1900's) saw the number of articles increase; however, during the early $21^{st}$ century (2000's) the number of published articles decreased slowly. The number of articles focusing on physiological and embryological research was significantly greater than taxonomic and ecological studies. Rainbow trout and Nile tilapia were the main research topic, specifically physiological and embryological research associated with the aquaculture of these species. Ecological studies were only conducted on the distribution and effect of large-mouth bass and nutria. The ecological risk associated with freshwater exotic species has been expressed yet the scientific information might be insufficient to remove doubt about ecological issues as expressed by interested by individuals and policy makers due to bias in research topics with respect to freshwater exotic species. The research topics of freshwater exotic species would have to diversify to effectively manage freshwater exotic species.

• Journal of Digital Convergence
• /
• v.17 no.5
• /
• pp.137-143
• /
• 2019

Fake news is a form of misinformation that has the advantage of rapid spreading of information on media platforms that users interact with, such as social media. There has been a lot of social problems due to the recent increase in fake news. In this paper, we propose a method to detect such false news. Previous research on fake news detection mainly focused on text analysis. This research focuses on a network where social media news spreads, generates qualities with DeepWalk, a network embedding method, and classifies fake news using logistic regression analysis. We conducted an experiment on fake news detection using 211 news on the Internet and 1.2 million news diffusion network data. The results show that the accuracy of false network detection using network embedding is 10.6% higher than that of text analysis. In addition, fake news detection, which combines text analysis and network embedding, does not show an increase in accuracy over network embedding. The results of this study can be effectively applied to the detection of fake news that organizations spread online.