• The Korean Journal of Applied Statistics
• /
• v.33 no.2
• /
• pp.171-184
• /
• 2020

Major League Baseball (MLB) records and releases the trajectory data for every baseball pitch, called the PITCHf/x, using three high-speed cameras installed in every stadium. In a previous study, the quality of the pitch was assessed as the expected number of bases yielded using PITCHf/x data. However, the number of bases yielded does not always lead to baseball scores, or runs. In this paper, we assess the quality of a pitch by combining baseball analytics metric Run Expectancy and Run Value using a Random Forests model. We compare the quality of pitches evaluated with Run Value to the quality of pitches evaluated with the expected number of bases yielded.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.601-603
• /
• 2012

본 논문에서는 의료영상의 특성을 반영하여 영상의 그래디언트 방향 값을 특징으로 하는 Oriented Center Symmetric Local Binary Patterns (OCS-LBP) 특징을 개발하고 추출된 특징 값에 대해 차원을 줄이고 의미 있는 특징 단위로 재 생성하기 위해 Bag-of-Feature (BoF)를 적용하였다. 검색을 위해서는 기존의 영상 검색 방법과는 다르게, 학습 영상을 이용하여 랜덤 포리스트 (Random Forest)를 사전에 학습시켜 데이터베이스 영상을 N 개의 클래스로 자동 분류 시키고, 질의로 입력된 영상을 같은 방법으로 랜덤 포리스트에 적용하여 상위 확률 값을 갖는 2 개의 클래스에서만 K-nearest neighbor 방법으로 유사 영상을 검색결과로 제시하는 새로운 영상검색 방법을 제시하였다. 실험결과에서 본 논문의 우수성을 증명하기 위해 일반적인 유사성 측정 방법과 랜덤 포리스트를 이용한 방법의 검색 성능 및 시간을 비교하였고, 검색 성능과 시간 면에서 상대적으로 매우 우수한 성능을 보여줌을 증명하였다.

• The KIPS Transactions:PartC
• /
• v.11C no.7 s.96
• /
• pp.917-922
• /
• 2004

Mobile ad hoc networks (MANETs) are networks of mobile nodes that have no fixed network infrastructure. Since the mobile node's location changes frequently, it is an attractive area to maintain the node's location efficiently. In this paper, we present an adaptive randomized database group (ARDG) scheme to manage the mobile nodes mobility in MHANETs. The proposed scheme stores the network nodes' location in location databases to manage the nodes' mobility. When a mobile node changes its location or needs a node's location, the node randomly select some databases to update or que교 the location information. The number of the selected databases is fixed in the case of querying while the number of the databases is determined according to the node's popularity in the case of updating. We evaluated the performance of the proposed scheme using an analytical model, and compared the performance with that of the conventional randomized database group (RDG) scheme.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.8
• /
• pp.165-170
• /
• 2024

This study aims to verify the accuracy of the air quality management system in Yangju City using an artificial intelligence (AI) evaluation model. The consistency and reliability of fine dust data were assessed by comparing public data from the Ministry of Environment with data from Yangju City's air quality management system. To this end, we analyzed the completeness, uniqueness, validity, consistency, accuracy, and integrity of the data. Exploratory statistical analysis was employed to compare data consistency. The results of the AI-based data quality index evaluation revealed no statistically significant differences between the two datasets. Among AI-based algorithms, the random forest model demonstrated the highest predictive accuracy, with its performance evaluated through ROC curves and AUC. Notably, the random forest model was identified as a valuable tool for optimizing the air quality management system. This study confirms that the reliability and suitability of fine dust data can be effectively assessed using AI-based model performance evaluation, contributing to the advancement of air quality management strategies.

• The Journal of Korean Association of Computer Education
• /
• v.18 no.5
• /
• pp.63-70
• /
• 2015

In the environment that the complex conditions need to be satisfied, skyline query have been applied to various field. To processing a skyline query in centralized scheme, several techniques have been suggested and recently map/reduce platform based approaches has been proposed which divides data space into multiple partitions for the vast volume of multidimensional data. However, the performances of these approaches are fluctuated due to the uneven data loading between servers and redundant tasks. Motivated by these issues, this paper suggests a novel technique called MR-DEAP which solves the uneven data loading using the random sampling. The experimental result gains the proposed MR-DEAP outperforms MR-Angular and MR-BNL scheme.

• Journal of Internet Computing and Services
• /
• v.18 no.5
• /
• pp.31-38
• /
• 2017

Recently with the rapid development of LBS (Location-based Services) technology, approaches of protecting user's location have gained tremendous attentions. For using LBS, users need to forward their real locations to LBS server. However, if the user sends his/her real location to LBS server, the server will have the all the information about user in LBS. Moreover, if the user opens it to LBS server for a long time, the trajectory of user may be released. In this paper, we propose GTC (Grid-based Trajectory Cloaking) method to address the privacy issue. Different from existing approaches, firstly the GTC method sets the predicting trajectory and divides the map into $2^n*2^n$ grid. After that we will generate cloaking regions according to user's desired privacy level. Finally the user sends them to LBS server randomly. The GTC method can make the cost of process less than sequential trajectory k-anonymity. Because of confusing the departure and destination, LBS server could not know the user's trajectory any more. Thus, we significantly improve the privacy level. evaluation results further verify the effectiveness and efficiency of our GTC method.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.22 no.10
• /
• pp.97-103
• /
• 2008

Recently, it is increased to apply sensorless drive for BLDC (Brushless DC) motor to household electrical appliances, especially in the refrigerators and air conditioners, to reduce the cost and the acoustic noise by the operation and to make their functions more comfortable for human beings. In this paper, low-cost sensorless drive for BLDE motor is implemented by random PWM (Pulse Width Modulation). The experimental results show that the electromagnetic noise was reduced and the sound quality was improved by BLDC motor sensorless random PWM Control.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.401-404
• /
• 2008

u-GIS 환경에서 발생하는 시공간 데이터는 지속적으로 발생하는 데이터 스트림의 특성을 갖으며, 그런 특성으로 인하여 데이터 발생량이 급격히 증가함에 따라 데이터 손실 및 시스템 성능 저하현상이 발생한다. 이를 해결하기 위해 부하 분산 연구들이 활발히 진행되어 오고 있다. 그러나 기존의 연구 방식인 랜덤 부하 분산 방식과 의미적 부하 분산 방식은 현 u-GIS 환경에서 부하 분산 속도 및 질의 결과의 정확도 측면에 만족스럽지 못한 결과를 준다. 그래서 본 논문에서는 우선순위를 이용한 차등적 부하 분산(DLSM : Different Load Shedding using MAP table)기법을 제안한다. DLSM 기법은 등록된 공간질의의 공간연산을 통해 영역의 우선순위를 미리 부여하고, 데이터가 발생하여 질의 처리기로 유입되기 전 우선순위를 파악한다. 데이터는 우선순위 단계에 따라 유입량을 확인 후 삭제 여부가 결정된다. 결과적으로 부하 분산 속도와 질의 결과의 정확도를 향상시켰다.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.5 no.2
• /
• pp.200-207
• /
• 2000

Recently, it is increased to apply the inverter system to household electrical appliances, especially in the air c conditioners, refrigerators and washing machines, to reduce the power consumption and the acoustic noise by t the low speed operation, and to make their functions more comfortable for human beings. For the inverter s systems, however, it is highly required to reduce the undesirable electromagnetic noise and psychoacoustic n noise generated by PWM for variable speed operation. In this paper, the electromagnetic noise for the d detenninistic PWM and random PWM for the reciprocating compressors driven by the brusWess dc motor was a analyzed. It was also verified through the experiment that the elt'Ct$\tau$omagnetic noise was reduced and the s sound quality was improved by applying the random PWM.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.340-343
• /
• 2009

u-GIS 환경에서 GeoSensor로부터 수집되는 시공간 데이터는 데이터 스트림의 특징을 포함한다. 데이터 스트림은 다양한 입력 속도로 끊임없이 입력되고, 데이터의 크기 또한 가변적이다. 이런 이유로 한정적인 메모리와 처리능력의 시스템은 과부하 현상이 발생한다. 이를 해결하기 위해 초과되는 데이터를 버려 메모리 초과를 방지하는 기법들이 연구되고 있다. 공간질의는 공간과 위치 값을 기반으로 이루어지는 연산으로 공간질의 정확도는 공간과 위치 정보를 통해 보장된다. 그러나 기존 기법인 랜덤부하제한 기법과 의미적부하제한 기법은 공간질의가 요구하는 공간과 위치 값에 대해 고려하지 않고 삭제하기 때문에 공간질의에 대한 정확도가 감소하는 문제를 갖는다. 본 논문에서는 공간의 이용도를 이용하여 차등적 비율을 적용한 부하제한 기법은 연구하였다. 이 기법은 등록된 공간질의의 영역 겹침 정도에 따라 중요 레벨을 증가시키고, 이를 토대로 시공간 데이터의 중요도를 파악하여 중요도마다 주어진 비율에 의하여 차등적으로 삭제한다. 결과적으로 기존 기법보다 다소 높은 Drop rate를 통해 질의 처리 속도를 빠르게 회복시켰으며, 중요 데이터를 최대한 유지하여 Error rate를 감소시켰다.