• 부동산연구
• /
• 제27권3호
• /
• pp.71-86
• /
• 2017

본 연구의 목적은 딥 러닝 방법을 부동산가격지수 예측에 적용해보고, 기존의 시계열분석 방법과의 비교를 통해 부동산 시장 예측의 새로운 방법으로서 활용가능성을 확인하는 것이다. 딥 러닝(deep learning)방법인 DNN(Deep Neural Networks)모형 및 LSTM(Long Shot Term Memory networks)모형과 시계열분석 방법인 ARIMA(autoregressive integrated moving average)모형을 이용하여 여러 가지 부동산가격지수에 대한 예측을 시도하였다. 연구결과 첫째, 딥 러닝 방법의 예측력이 시계열분석 방법보다 우수한 것으로 나타났다. 둘째, 딥 러닝 방법 중에서는 DNN모형의 예측력이 LSTM모형의 예측력보다 우수하나 그 정도는 미미한 수준인 것으로 나타났다. 셋째, 딥 러닝 방법과 ARIMA모형은 부동산 가격지수(real estate price index) 중 아파트 실거래가격지수(housing sales price index)에 대한 예측력이 가장 부족한 것으로 나타났다. 향후 딥 러닝 방법을 활용함으로써 부동산 시장에 대한 예측의 정확성을 제고할 수 있을 것으로 기대된다.

• 한국문헌정보학회지
• /
• 제56권2호
• /
• pp.109-132
• /
• 2022

본 연구는 공공도서관에서 운영 되었던 독서마라톤 대회의 현황을 비교 분석하여 성과 향상을 위한 과제를 탐색하고, 더불어 향후 이 대회를 적용할 곳들을 위한 가이드라인을 제공하는데 목적이 있었다. 따라서 연구 목적 달성을 위해 코로나-19 팬데믹 이전부터 2021년까지 독서마라톤 대회를 운영한 공공도서관 다섯 곳의 현황을 수집하여 그 내용들을 비교 분석하였다. 그 결과 다섯 곳의 공공도서관이 독서마라톤 대회를 운영한 평균 기간은 10년이었고, 한 해 동안 대회 기간은 평균 5.87개월이었다. 또한 완주 비율이 가장 높은 대상은 아동이었으며, 완주에 대한 보상으로는 증서를 발급하고 대출권수를 상향시켜 주는 경우가 가장 많았다. 그러나 대회 담당자들은 신청자 확보 및 증대, 도달한 거리 산정 등에 대한 어려움을 호소하고 있어, 분석한 내용을 바탕으로 독서마라톤 대회 성과 향상을 위한 개선 과제를 도출하고 그에 대한 해결 방안도 제언하였다.

• 한국통신학회논문지
• /
• 제39C권10호
• /
• pp.909-919
• /
• 2014

License plate (LP) detection is the most imperative part of an automatic LP recognition (LPR) system. Typical LPR contains two steps, namely LP detection (LPD) and character recognition. In this paper, we propose an efficient Vehicle-to-LP detection framework which combines with an adaptive GMM (Gaussian Mixture Model) and a cascade of boosted classifiers to make a faster vehicle LP detector. To develop a background model by using a GMM is possible in the circumstance of a fixed camera and extracts the motions using background subtraction. Firstly, an adaptive GMM is used to find the region of interest (ROI) on which motion detectors are running to detect the vehicle area as blobs ROIs. Secondly, a cascade of boosted classifiers is executed on the blobs ROIs to detect a LP. The experimental results on our test video with the resolution of $720{\times}576$ show that the LPD rate of the proposed system is 99.14% and the average computational time is approximately 42ms.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2011년도 정기총회 및 추계학술대회 논문집
• /
• pp.1739-1742
• /
• 2011

Public transportation network in our country is concentrated and advanced focusing on urban area in order to secure economic feasibility. As a result, as dependence on private vehicles is relatively getting higher in public transportation vulnerable area, traffic problem occurs since the average running speed in urban area is 22.5km/h. This paper has an objective to suggest an improvement plan by selecting public transportation vulnerable area, and defining according to urban structure, formation and function, and understand traffic characteristics and draw problems. As a method selecting public transportation vulnerable area, an area with high division rate of vehicle was selected as a primary proposed site by calculating division rate of means of public transportation according to area. Final proposed site was selected by using GIS Buffering technique aiming at selected proposed site, and selecting non-benefit area 500m outside, which is the road limit distance from each subway and bus station. Lastly, the degree of improvement effect was studied by constructing imaginary public transportation network aiming at final proposed site and comparing to the amount of change in division rate of means of transportation.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 하계종합학술대회 논문집(1)
• /
• pp.175-178
• /
• 2004

This paper describes a new IP address lookup algorithm using a binary search on multiple balanced trees stored in one memory. The proposed scheme has 3 different tables; a range table, a main table, and multiple sub-tables. The range table includes $2^8$ entries of 22 bits wide. Each of the main table and sub-table entries is composed of fields for a prefix, a prefix length, the number of sub-table entries, a sub-table pointer, and a forwarding RAM pointer. Binary searches are performed in the main table and the multiple sub-tables in sequence. Address lookups in our proposed scheme are achieved by memory access times of 11 in average, 1 in minimum, and 24 in maximum using 267 Kbytes of memory for 38.000 prefixes. Hence the forwarding table of the proposed scheme is stored into L2 cache, and the address lookup algorithm is implemented in software running on general purpose processor. Since the proposed scheme only depends on the number of prefixes not the length of prefixes, it is easily scaled to IPv6.

• 응용통계연구
• /
• 제29권7호
• /
• pp.1271-1281
• /
• 2016

야구에서 타자의 능력을 측정하는 많은 세이버메트릭스 통계량들 중에서 대체선수대비승수(wins above replacement; WAR)은 가장 많이 쓰이는 통계량이다. WAR은 선수의 공격능력과 주루능력, 수비능력 등을 하나의 수치로 표현하는 방법이란 점에서 큰 장점을 가지고 있다. 본 논문에서는 지난 3년간(2013-2015년) 한국프로야구 기록 자료를 바탕으로 세이버메트릭스 변수들의 값을 구한 뒤, 이를 이용하여 WAR을 대체할 수 있는 타자능력지수를 제안하였다. 타자능력지수는 산술평균방법, 가중평균방법, 주성분회귀분석 등을 통해 산출하고 WAR과 비교하여 가장 관계가 높은 방법을 선택하였다.

• 한국안전학회지
• /
• 제31권5호
• /
• pp.158-170
• /
• 2016

It has been applied several ways in order to prevent traffic accidents and speed management on the road which is one of very important factors to prevent accidents. In this study, it is analyzed whether the facility such as the flashing and crosswalk, speed cameras and intermittent warning signs affect on a driving deceleration or not. It has been shown that the flashing and crosswalk and speed cameras except intermittent warning signs play a role of reducing the driving speed. The result showed that these traffic safety facilities help to improve the average speed as well as velocity distribution which was derived by the speed deviation even if there were the difference between facilities. Speed limit display does not affect the deceleration of the driver. Therefore, it was founded in this research that there were the driving speed reduction effect due to the facilities on the road. In order to control the driving speed on the road, it is necessary to install facilities with appropriate location and spacing on the road. In addition, place that turns out suppressed speeding through speed management also are required adequate facilities such as speed cameras.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2003년도 제2차 연례학술대회 발표논문집
• /
• pp.133-138
• /
• 2003

The prediction of protein secondary structure has been an important bioinformatics tool that is an essential component of the template-based protein tertiary structure prediction process. It has been known that the predicted secondary structure information improves both the fold recognition performance and the alignment accuracy. In this paper, we describe several novel ideas that may improve the prediction accuracy. The main idea is motivated by an observation that the protein's structural information, especially when it is combined with the evolutionary information, significantly improves the accuracy of the predicted tertiary structure. From the non-redundant set of protein structures, we derive the 'potential' parameters for the protein secondary structure prediction that contains the structural information of proteins, by following the procedure similar to the way to derive the directional information table of GOR method. Those potential parameters are combined with the frequency matrices obtained by running PSI-BLAST to construct the feature vectors that are used to train the support vector machines (SVM) to build the secondary structure classifiers. Moreover, the problem of huge model file size, which is one of the known shortcomings of SVM, is partially overcome by reducing the size of training data by filtering out the redundancy not only at the protein level but also at the feature vector level. A preliminary result measured by the average three-state prediction accuracy is encouraging.

• 전기학회논문지
• /
• 제63권3호
• /
• pp.395-403
• /
• 2014

This paper presents a new approach to improve the signal to noise ratio (SNR) for local seismic disaster preventing system in densely populated area. The seismic data measured in the local site includes various sensing noises (offset or measurement noise) and man-made/natural noises (road and rail traffic noise, rotating or hammering machinery noise, human activity noise such as walking and running, wind/atmospheric pressure-generated noise, etc.). These additive noises are different in time and frequency characters. The proposed method uses 3-stages processing to reduce these different additive noises. In the first stage, misalignment offset noise are diminished by time average processing, and then the second and third stages, coherent/incoherent noises such as man-made/natural noises are suppressed by array stacking. In addition, we derived the theoretical equation of the SNR gain improved by the proposed method. To evaluate the performance of the proposed method, computer simulations were performed with real seismic data and test equipment generated data as the input.

• Journal of Computing Science and Engineering
• /
• 제9권1호
• /
• pp.1-8
• /
• 2015

Accessing and searching in a sequence of numbers are fundamental operations in computing that are encountered in a wide range of applications. One of the applications of the problem is cryptanalytic time-memory tradeoff which is aimed at a one-way function. A rainbow table, which is a common method for the time-memory tradeoff, contains elements from an input domain of a hash function that are normally sorted integers. In this paper, we present a practical indexing method for a monotonically increasing static sequence of numbers where the access and search queries can be addressed efficiently in terms of both time and space complexity. For a sequence of n numbers from a universe $U=\{0,{\ldots},m-1\}$, our data structure requires n lg(m/n) + O(n) bits with constant average running time for both access and search queries. We also give an analysis of the time and space complexities of the data structure, supported by experiments with rainbow tables.