• 정보보호학회논문지
• /
• 제25권5호
• /
• pp.1131-1141
• /
• 2015

본 논문에서는 MMORPG에서 각 캐릭터의 소지금 증가/감소 이벤트 로그 데이터를 위주로 플레이어의 액션 로그 데이터를 조사하여 게임봇을 탐지하는 기계 학습 기반의 새로운 접근 방법을 제안한다. 게임봇 계정과 일반 계정을 구분하는 주요 피쳐를 추출하기 위해 밀도 기반 군집화 알고리즘의 하나인 DBSCAN (Density Based Spatial Clustering of Application with Noise)를 이용하였다. DBSCAN 알고리즘을 통해 각 플레이어의 소지금 증가/감소 위치 좌표를 클러스터링하고, 그 결과 생성된 클러스터의 수, 코어 포인트의 비율, 멤버 포인트의 비율, 노이즈 포인트의 비율과 같은 공간적 특성을 나타내는 값들을 추출하였다. 해당 피쳐들을 사용하면 게임봇 개발자들이 게임봇 탐지 시스템의 원리를 알더라도 넓은 지역을 돌아다니며 사냥을 하도록 게임봇 프로그램을 제작하는 것은 매우 비효율적이기 때문에 탐지 시스템을 우회하기 어렵게 된다. 결과적으로, 게임봇은 소지금 변동 좌표 데이터로부터 추출한 공간적 특성에서 일반유저와 명확한 차이를 보였다. 예를 들면, DBSCAN 클러스터링 결과 중 노이즈 포인트의 비율에서 게임봇은 5% 이하의 낮은 값을 가지는 반면에 일반 유저들은 대부분 높은 값을 갖는다. 실제 MMORPG의 액션 로그 데이터를 이용한 게임봇 탐지에서, 본 논문에서 제안된 시스템은 높은 탐지율의 우수한 성능을 보였다.

• 한국산업융합학회 논문집
• /
• 제7권3호
• /
• pp.281-288
• /
• 2004

The objective of this study is to propose a methodology of the flood runoff analysis in steep mountainous basins and the analysis basin is the Jasa valley basin in Chungju city Analyzing the spatial pattern of the rainfall in 1994. 6 30~7.1, the seasonal rainy front was tied up in the whole central district, and the rainfall center was moving from the northern Chungbuk province to the northern Kyongbuk province and caused heavy storm. Analyzing the temporal pattern with the Huff method, the 52.5% of the rainfall was concentrated on the 3rd quartile. Rainfall frequency analysis is accomplished by five distribution types; 2-parameter Lognomal, 3-parameter Lognomal, Pearson Type III, Log-Pearson Type III and Extremal Type I distribution Rainfall-runoff analysis in Jasa valley basin was made using HEC-HMS model. Jasa valley basin was divided into 3 sub-basins and the analysis point was 3 points{A, B and C point) With the rainfall data measured by the 10 minutes, the flood runoff also was calculated by as many minutes. SCS CN model, Clark UH model and Muskingum routing model in HEC-HMS model were used to simulate the runoff volume using selected rainfall event.

• Journal of Electrical Engineering and Technology
• /
• 제6권6호
• /
• pp.883-887
• /
• 2011

Location-based service (LBS) is becoming an important part of the information technology (IT) business. Localization is a core technology for LBS because LBS is based on the position of each device or user. In case of outdoor, GPS - which is used to determine the position of a moving user - is the dominant technology. As satellite signal cannot reach indoor, GPS cannot be used in indoor environment. Therefore, research and study about indoor localization technology, which has the same accuracy as an outdoor GPS, is needed for "seamless LBS". For indoor localization, we consider the IEEE802.11 WLAN environment. Generally, received signal strength indicator (RSSI) is used to obtain a specific position of the user under the WLAN environment. RSSI has a characteristic that is decreased over distance. To use RSSI at indoor localization, a mathematical model of RSSI, which reflects its characteristic, is used. However, this RSSI of the mathematical model is different from a real RSSI, which, in reality, has a sensitive parameter that is much affected by the propagation environment. This difference causes the occurrence of localization error. Thus, it is necessary to set a proper RSSI model in order to obtain an accurate localization result. We propose a method in which the parameters of the propagation environment are determined using only RSSI measurements obtained during localization.

• Water Engineering Research
• /
• 제6권1호
• /
• pp.31-38
• /
• 2005

Groundwater recharge is defined in an addition of water to groundwater reservoir. Recently, many people have been moving to the Edwards aquifer and urban and agricultural industry have been expending. Hydrologists and water planning managers concern about insufficient groundwater amounts and irrigation water price variability. In this paper, I focus on estimates of local recharge volumes and quantify preferential flow through GIS technique. Chloride Mass Balance (CMB) and hydrochemical components have been widely applied to recharge rate and evaluate flow paths. The CMB method is based on relationship between wet-dry chloride deposition data and Rainfall data. These data are manipulated using ArcGIS. Especially, hydrochemical concentration distribution is good index for groundwater residence times or flow paths such as $[Mg^{2+}]/[Ca^{2+}],[Cl]$ and log$([Ca^{2+}]+[Mg^{2+}])/[Na^+]$. Well information such as hydrological-hydrochemical data are imported into ArcGIS and manipulated by interpolation techniques. For each potentiometric surface and water quality, point data are converted to spatial data through each Kriging and Inverse Distance Weighted (IDW) techniques.

• 전자공학회논문지B
• /
• 제28B권7호
• /
• pp.566-577
• /
• 1991

In this paper, the method of image segmentation based on a pyramid of reduced resolution versions of the input input image is persented. In a pyramid structure, two regions (a given pixel and its mother pixels) are compared by the proposed inverse-mother-son variance ratio (IMSVR) method for the detection of an optinal object pixel and are determined whether they are similar enough to be viewed as one region or disparate to be viewed as ditinct regions By the proposed method, an l`timal object pixel has been setectedat some level, it is necessary to retrieve its boundary precisely. Moving down the pyramid to levels of higher resolution is requires. In this paper, the top-sown pyramid traversing algorithm for an image segmentation using a pyrmid structure is presented. Using the computer simulation, the results by the proposed statistical method and object traversing method are investigated for the binary image and the real image at the results of computer simulation, the proposed method of image segmentation based on a pyramid structure seem to have useful properties and deserve consideration as a possible alternative to existing methods of omage segmentation. The computation for the proposed method is required 0 (log n), for an TEX>$n{\times}n$ input image.

• Journal of Forest and Environmental Science
• /
• 제36권1호
• /
• pp.7-16
• /
• 2020

Tree size distribution modelling is an integral part of forest management. Most distribution yield systems rely on some flexible probability models. In this study, a simple finite mixture of two components two-parameter Weibull distribution was compared with complex four-parameter distributions in terms of their fitness to predict tree size distribution of teak (Tectona grandis Linn f) plantations. Also, a system of equation was developed using Seemingly Unrelated Regression wherein the size distributions of the stand were predicted. Generalized beta, Johnson's SB, Logit-Logistic and generalized Weibull distributions were the four-parameter distributions considered. The Kolmogorov-Smirnov test and negative log-likelihood value were used to assess the distributions. The results show that the simple finite mixture outperformed the four-parameter distributions especially in stands that are bimodal and heavily skewed. Twelve models were developed in the system of equation-one for predicting mean diameter, seven for predicting percentiles and four for predicting the parameters of the finite mixture distribution. Predictions from the system of equation are reasonable and compare well with observed distributions of the stand. This simplified mixture would allow for wider application in distribution modelling and can also be integrated as component model in stand density management diagram.

• 제어로봇시스템학회논문지
• /
• 제22권7호
• /
• pp.562-569
• /
• 2016

This paper proposes a navigation system with a robust localization method for an underwater unmanned vehicle. For robust localization with IMU (Inertial Measurement Unit), a DVL (Doppler Velocity Log), and depth sensors, the EKF (Extended Kalman Filter) has been utilized to fuse multiple nonlinear data. Note that the GPS (Global Positioning System), which can obtain the absolute coordinates of the vehicle, cannot be used in the water. Additionally, the DVL has been used for measuring the relative velocity of the underwater vehicle. The DVL sensor measures the velocity of an object by using Doppler effects, which cause sound frequency changes from the relative velocity between a sound source and an observer. When the vehicle is moving, the motion trajectory to a target position can be recorded by the sensors attached to the vehicle. The performance of the proposed navigation system has been verified through real experiments in which an underwater unmanned vehicle reached a target position by using an IMU as a primary sensor and a DVL as the secondary sensor.

• Management Science and Financial Engineering
• /
• 제11권3호
• /
• pp.95-107
• /
• 2005

As online shopping malls have emerged as a substantial shopping channel, they have used various sales promotion strategies to acquire new customers. Most of these strategies have been applied by offline malls for years. One, loss-leader pricing, is a type of promotional pricing in which stores sell well known products below their marginal cost, in order to attract customers and induce them to purchase more goods through impulse buying. This strategy is based on the expectation that customers will factor transaction costs into their purchasing decisions. However, its application to online malls fails to recognize that transaction costs are lower online, and that customers will behave differently as a result. Our study predicts that loss-leader pricing will not work online because online malls entail lower searching and moving costs than offline malls The study examines the effectiveness of loss-leader pricing with empirical data from a survey as well as log data from a Korean online shopping mall. The results show that while loss-leader pricing does attract customers to online shopping malls, it encourages cherry-picking rather than impulse purchases of regular-price goods.

• 한국음향학회지
• /
• 제32권2호
• /
• pp.147-156
• /
• 2013

본 논문에서는 이동하는 로봇에 탑재한 대화체 음성인식기의 주위 잡음 환경에 강인한 새로운 음성 끝점 검출 기법을 제안한다. 기존의 기법은 특징 값의 갑작스러운 변화점을 찾기 위해 에지 검출 필터(edge detection filter)를 적용하여 끝점을 찾았다. 하지만 프레임 에너지의 특징은 잡음 환경에서 불안정하기 때문에 음성의 끝점을 정확하게 찾기 어렵다. 그러므로 두 번의 고속 퓨리에 변환과 통계적 모델 기반의 특징 추출 기법을 제안하여 에지 검출 필터에 적용한다. 제안한 기법이 기존의 기법보다 강인한 특징이 될 수 있음을 본 실험을 통하여 확인하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 제38회 하계학술대회
• /
• pp.2040-2041
• /
• 2007

In this paper, the UHF PD(Partial Discharge) sensors for $SF_6$ insulated MV $SF_6$ switchgear have been proposed and related investigations have been performed in order to detect the PD which were produced inside the MV $SF_6$ switchgear. Firstly, the internal type UHF PD sensor based on spiral antenna theory has been developed. This type sensor is highly sensitive and has lowly effect on by on-site noise. Secondly, the external type UHF PD sensor was developed based on log periodic antenna concept. This type sensor is removable and detectable for operating switchgear. These sensors were designed and simulated using RF simulation tool. In order to verify the sensitivity of these sensors, we performed the on-site test using the mock-up switchgears including the artificial defects which were the protrusion on high voltage conductor, free moving metal particle and surface defect on insulator. These mock-up switchgear were installed on the test distribution line.