http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
스마트폰 가속도 센서를 이용한 강건한 사용자 행위 인지 방법
전명중 ( Jeon Myung Joong ),박영택 ( Park Young Tack ) 한국정보처리학회 2013 정보처리학회논문지. 소프트웨어 및 데이터 공학 Vol.2 No.9
Recently, with the advent of smart phones, it brought many changes in lives of modern people. Especially, application utilizing the sensor information of smart phone ,which provides the service adapted by user situations, has been emerged. Sensor data of smart phone can be used for recognizing the user situation, Because it is closely related to the behavior and habits of the user. currently, GPS sensor one of mobile sensor has been utilized a lot to recognize basic user activity. But, depending on the user situation, activity recognition system cannot receive GPS signal, and also not collect received data. So utilization is reduced. In this paper, for solving this problem, we suggest a method of user activity recognition that focused on the accelerometer sensor data using smart phone. Accelerometer sensor is stable to collect the data and it`s sensitive to user behavior. Finally this paper suggests a noble approach to use state transition diagrams which represent the natural flow of user activity changes for enhancing the accuracy of user activity recognition.
Extracting Rules from Neural Networks with Continuous Attributes
Batselem Jagvaral(바트셀렘),Wan-Gon Lee(이완곤),Myung-joong Jeon(전명중),Hyun-Kyu Park(박현규),Young-Tack Park(박영택 ) Korean Institute of Information Scientists and Eng 2018 정보과학회논문지 Vol.45 No.1
Over the decades, neural networks have been successfully used in numerous applications from speech recognition to image classification. However, these neural networks cannot explain their results and one needs to know how and why a specific conclusion was drawn. Most studies focus on extracting binary rules from neural networks, which is often impractical to do, since data sets used for machine learning applications contain continuous values. To fill the gap, this paper presents an algorithm to extract logic rules from a trained neural network for data with continuous attributes. It uses hyperplane-based linear classifiers to extract rules with numeric values from trained weights between input and hidden layers and then combines these classifiers with binary rules learned from hidden and output layers to form non-linear classification rules. Experiments with different datasets show that the proposed approach can accurately extract logical rules for data with nonlinear continuous attributes.