A5087746362- Innovation in Digital Healthcare Systems
- Advanced Neural Network Applications
- Image Enhancement Techniques
- Technology and Data Analysis
- Text and Document Classification Technologies
- Advanced Text Analysis Techniques
- Autonomous Vehicle Technology and Safety
- Advanced Decision-Making Techniques
- Digital Transformation in Industry
- Traffic Prediction and Management Techniques
- Internet of Things and Social Network Interactions
- Video Surveillance and Tracking Methods
- Visual Attention and Saliency Detection
- Web Data Mining and Analysis
- Consumer Perception and Purchasing Behavior
- Education and Learning Interventions
- Time Series Analysis and Forecasting
- Diverse Approaches in Healthcare and Education Studies
- Data Mining Algorithms and Applications
- Advanced Machining and Optimization Techniques
- Industrial Vision Systems and Defect Detection
- Sentiment Analysis and Opinion Mining
- Data Management and Algorithms
Kyonggi University
2019-2022
Korean people are exposed to stress due the constant competitive structure caused by rapid industrialization. As a result, there is need for ways that can effectively manage and help improve quality of life. Therefore, this study proposes an activity recommendation model using rank correlation chronic management. Using Spearman’s coefficient, proposed finds correlations between users’ Positive Activity Stress Management (PASM), Negative (NASM), Perceived Scale (PSS). coefficient improves...
This study proposes the prefixSpan based pattern mining using time sliding weight from streaming data. To discover sequential patterns, it applies a to create structure of projected DB Tree. For weight, window is applied data calculate label and support window. When Tree designed, calculated for each inserted in table. At this time, tree updated by deleting node whose less than reference value. reason, whenever updated, sorted again. The reordering process removes influence applying weights....
With the increase in traffic accident rates, risk detection is becoming increasingly important. Moreover, it necessary to provide appropriate information considering user locations and routes design an analysis method accordingly. This paper proposes a word-embedding-based document classification model for detecting emerging risks using quantity termed sentiment similarity weight (SSW). The proposed detects by classifying importance polarity of keywords document. Conventional methods fail...
High-quality, large-capacity data are essential for training a deep learning vision model. However, to construct crop image data, absolute growth time is required growth. In addition, it characterized by unbalanced with fewer abnormal than normal data. Therefore, building high-quality, large-scale datasets challenging. Many studies have used augmentation of plant images solve this problem. plants require that does not compromise their color, texture, or shape. This study proposes the use...
현대 사회에서 성범죄율은 지속적으로 증가하고 있다. 특히 성범죄자의 경우 재범률이 높기 때문에 재발을 막기 위한 제도적 혹은 사회적 필요성이 점차 이를 위해 본 논문에서 범죄예방을 CCTV 영상 기반의 실시간 범죄자 안면 인식 시스템을 제안한다. 이는 인공지능 안면인식을 통해 성범죄 예방 및 재범 방지를 일련의 프로세스다. 하지만, 기술을 실질적으로 활용하기에는 문제점이 시스템은 영상에서 안면을 찾는 탐지 단계와 탐지된 안면과의 일치도를 계산하는 매칭 단계로 이루어지며 단일 작업에 비해 큰 연산 비용이 요구된다. 연구에서는 탐지할 때 벡터화 작업을 처리 시간의 소모를 줄여 작업 시에 비용을 줄이고, 이후 영상의 크기를 늘리는 AB-box 방식을 사용해 더 높은 정확도를 보인다.