A5101709029- Video Analysis and Summarization
- Multimedia Communication and Technology
- Peer-to-Peer Network Technologies
- Advanced Image and Video Retrieval Techniques
- User Authentication and Security Systems
- Cloud Computing and Resource Management
- Software System Performance and Reliability
- Network Security and Intrusion Detection
- Spam and Phishing Detection
- Privacy, Security, and Data Protection
- Speech Recognition and Synthesis
- Video Coding and Compression Technologies
- Web Data Mining and Analysis
- Medical Research and Treatments
- Geological Modeling and Analysis
- Smart Grid and Power Systems
- Caching and Content Delivery
- Energy and Environmental Systems
- Technology and Security Systems
- Advanced Authentication Protocols Security
- Data Mining Algorithms and Applications
- Recommender Systems and Techniques
- Advanced Text Analysis Techniques
- ECG Monitoring and Analysis
- Consumer Perception and Purchasing Behavior
Kookmin University
2024
Soongsil University
2009-2022
Jeonbuk National University
2008-2019
In this paper, a novel method is proposed to build an ensemble of classifiers by using feature selection schema. The schema identifies the best sets that affect arrhythmia classification. Firstly, number subsets are extracted applying original dataset. Then classification models built each subset. Finally, we combine adopting voting approach form ensemble. in our involves both error rate and calculate score classifier method, depends on extracting order subsets. experiment, applied dataset...
This paper proposes a new incremental weighted mining based on RFM ((Recency, Frequency, Monetary) analysis for recommending prediction in u-commerce.Association rules search the associated item set large database.Association are frequently used by marketing pattern e-commerce, recommendation to promote selling product marketing.The proposing method can extract frequent items and create association using rapidly when data added persistently order predict changing trends emphasizing important...
Technologies based on the internet are improving continuously and so harmful websites such as pornography or illegal gambling websites. In addition, it is characteristics of that changes made to web address its contents take effect almost instantaneously. Therefore, not easy identify from those not. There two ways make decision: manual examination automated searching for certain texts, videos, sounds. These methods require a lot time. this paper we propose method identifying by analyzing...
This study aims to analyze and improve the auditory user experience of practitioners subjects in use aesthetic medical devices.In particular, focused on lifting devices, which have a large number touch points treatment environment.
Today, the interest in intellectual property rights is increasing, and importance of patents, one properties, especially growing. Because patents include a lot technical information such as paper, many things could be obtained through patent analysis. However, large amount data number continues to grow. In addition, each country provides different formats most them are semi-structured or unstructured so it not easy analyze them. Therefore, this big data-based analysis system proposed these...
The recent proliferation of satellite imaging technologies and advancements internet infrastructure have encouraged provisions image services in web environments. However, most these still rely upon low-resolution images combined with DEM. In this paper, we propose the adaptive rendering engine which renders high-spatial resolution for proficient streaming on web. From Mt. Bukhan data as a pilot study, it is shown that proposing approach efficiently solves graphical issues real time...
This paper proposes an efficient storage virtualization system which allows users to view hard disk resources constituted of numerous nodes as a large logical space using distributed hash tables P2P networking. The proposed is developed at the device level operating (e.g. MS-Windows) and suitable for in Intranet environments. be recognized single Windows Explorer users' conveniences does not require any supplementary client program application layer. In addition, it enhances security via...
The advancement of information technology along with the proliferation communication and multimedia, have increased demand digital contents. Video data contents such as VOD (video on demand), NOD (news library, IPTV (Internet protocol television), UCC (user created contents) are getting more permeated in various application fields. sequential characteristic besides providing spatial temporal its 3D format, making searching or browsing ineffective due to long turnaround time.In this paper, we...
논문은 실시간 돌발 상황 정보 인식이 어려운 터널 내에서 CCTV를 통해 실시간으로 상황을 인식할 수 있는 시스템을 개발하였다. 상황은 차량의 역주행 및 정지, 사람의 보행, 화재 낙하물의 총 5가지 상황으로 정의하였다. 효율적으로 인식하기 위해 화재와 낙하물은 배경제거 알고리즘을 활용하고, 차량 사람은 YOLOv3를 기반으로 한 Deep-sort 활용하는 하이브리드 방식의 영상인식 기술을 사용하였다. 각 상황에 대한 175번의 부여하여 실험을 진행하였고, 이 중 160번의 돌발상황을 인식하여 91%의 정확도를 보였다.