• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.268-273
• /
• 2017

Recently, deep learning encoder-based approach has been actively applied in the field of sentiment classification. However, Long Short-Term Memory network deep learning encoder, the commonly used architecture, lacks the quality of vector representation when the length of the documents is prolonged. In this study, for effective classification of the sentiment documents, we suggest the use of attention method-based deep learning encoder that generates document vector representation by weighted sum of the outputs of Long Short-Term Memory network based on importance. In addition, we propose methods to modify the attention method-based deep learning encoder to suit the sentiment classification field, which consist of a part that is to applied to window attention method and an attention weight adjustment part. In the window attention method part, the weights are obtained in the window units to effectively recognize feeling features that consist of more than one word. In the attention weight adjustment part, the learned weights are smoothened. Experimental results revealed that the performance of the proposed method outperformed Long Short-Term Memory network encoder, showing 89.67% in accuracy criteria.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.6
• /
• pp.28-39
• /
• 2010

The study aims at offering a solution to the problems of transmission delay and data throughput decrease as the number of contending On-Board Units (OBU) increases by applying CSMA medium access control protocol based upon IEEE 802.11p. In a competition-based medium, contention probability becomes high as OBU increases. In order to improve the performance of this medium access layer, the author proposes EDCA which a adaptive adjustment of the Contention Windows (CW) considering traffic density and data type. EDCA applies fixed values of Minimum Contention Window (CWmin) and Maximum Contention Window (CWmax) for each of four kinds of Access Categories (AC) for channel-specific service differentiation. EDCA does not guarantee the channel-specific features and network state whereas it guarantees inter-AC differentiation by classifying into traffic features. Thus it is not possible to actively respond to a contention caused by network congestion occurring in a short moment in channel. As a solution, CWminAS(CWmin Adaptation Scheme) and ACATICT(Adaptive Contention window Adjustment Technique based on Individual Class Traffic) are proposed as active CW control techniques. In previous researches, the contention probabilities for each value of AC were not examined or a single channel based AC value was considered. And the channel-specific demands of IEEE 802.11p and the corresponding contention probabilities were not reflected in the studies. The study considers the collision number of a previous service section and the current network congestion proposes a dynamic control technique ACCW(Adaptive Control of Contention windows in considering the WAVE situation) for CW of the next channel.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.106.5-106
• /
• 2002

1. Introduction 2. Artificial Immune Networks and Navigation Algorithm 3. Obstacle Avoidance and Goal Approach Behavior 4. Weights Adjustment Using Neural Network 5. Velocity Control and Local Minimum Avoidance 6. Simulation 7. Conclusion

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.5 no.4
• /
• pp.121-127
• /
• 1985

This paper deals with simultaneous geodetic networks adjustment by geographical coordinates(${\varphi}$, ${\lambda}$). The adjustment computation is performed by variation of coordinates, and the classical method with fixed points and free networks are also compared. Provisional values for observation equations are computed by extended Gauss-mid lattitude formula using existing official coordinates. Bessel ellipsoid and unit weight are adopted. The processing of a test-network by distances yields the average root mean square error of position 6.2 cm for classical method and 2.4cm for free networks. The standard error of unit weight in a test-network is $1.66{\times}10^{-6}$ radian (0.3"), and the analysis of error ellipses shows that free networks are more normally distributed errors.

• Journal of Cadastre & Land InformatiX
• /
• v.45 no.2
• /
• pp.117-130
• /
• 2015

A least squares method for adjusting the horizontal network satisfies the conditions which is minimizing the sum of the squares of errors based on probability theory. This research compared accuracy of 3rd cadastral control points adjusted by traditional and least square method with respect to the result of Network-RTK. Test results showed the least square method more evenly distribute closure error than traditional method. Mean errors of least square and traditional adjusting method are 2.7cm, 2.2cm respectively. In addition, blunder in angle observations can be detected by comparing position errors which calculated by forward and backward initial coordinates. However, distance blunder cannot offer specific observation line occurred mistake because distance error propagates several observation lines which have similar directions.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.11 no.4
• /
• pp.133-141
• /
• 1991

The empirical method is used to integration adjustment for the coordinates revision of a national control point but the existing values are not to be changed or changed with small variation by suitable datum selection (for example, fixed points). This paper treats the modified weighted station parameter adjustment by quasi-observations, and the method used only variance elements of existing coordinates which is substituted for all covariance elements. The movement detection of unstable points and the junction adjustment of new networks are successfully executed by the method, in integration of new secondary networks to old-secondary-triangulation points which are in the absence of the original observations in Korea. The investigation results reveal that the accuracy of old-secondary-triangulation points is ${\pm}16^{{\prime}{\prime}}$(${\pm}0.08m$), which results from the densification of test network and the analyses of old survey specifications. and is ${\pm}2.3^{{\prime}{\prime}}$ in fixing of old-secondary-triangulation points.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.19 no.3
• /
• pp.283-290
• /
• 2001

This paper demonstrates the performance of industrial precise measurement using the digital close-range photograrmmetric system based on a off-the-shelf CCTV-type CCD camera. The system was constructed with a CCD camera and a PC with a frame grabber, coupled with digital image mensuration and self-calibrating bundle adjustment techniques. An artificial fish reef with cubic shape was taken as an object for the application test of the system and the digital images were acquired on multi-station convergent network around the object. The geometric calibration of the CCD camera and the phototriangulation of the entire surface of the object was carried out simultaneously by means of self-calibrating bundle adjustment technique. Also the system comprising a high resolution still-video camera Kodak DCS, which high accuracy potential has been already established, were employed in similar network condition. Then the results from two different camera systems were compared in the accuracies of phototriangulation.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.11-16
• /
• 2005

Recently, there are a lot of endeavor to implement a fuzzy theory to artificial neural network. Goh proposed the fuzzy single layer perceptron algorithm and advanced fuzzy perceptron based on the generalized delta rule to solve the XOR Problem and the classical Problem. However, it causes an increased amount of computation and some difficulties in application of the complicated image recognition. In this paper, we propose an enhanced fuzzy single layer Perceptron using the dynamic adjustment of threshold. This method is applied to the XOR problem, which used as the benchmark in the field of pattern recognition. The method is also applied to the recognition of digital image for image application. In a result of experiment, it does not always guarantee the convergence. However, the network show improved the learning time and has the high convergence rate.

• Journal of Energy Engineering
• /
• v.25 no.4
• /
• pp.226-235
• /
• 2016

This study intends to present policy proposals to make the transition from policy adjustment system based on hierarchial authority to deal with energy policy to network establishment based on the central government, local government, industry and civic group, that can be converted into governance. To this end, the legal foundation to make the network-based policy adjustment for government's energy policy possible first. Second, contribution should be made to establish governance related to central government's energy policy. Third, contribution should be made to establish governance related to local government's energy policy. Based on this, this study intends to secure policy measures to establish and improve governance related to energy policy in Korea.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.16 no.1
• /
• pp.85-94
• /
• 1998

In year 1996, the National Geography Institute(NGI) carried out a National Fundamental Network survey. A total of 220 baselines between 31 stations were occupied that the baseline length is 40 km-120 km(mean 67.4 km). A minimally constrained network adjustment with three dimensional baseline vectors, was carried out holding geocentric ITRF94 coordinates of the station SUWON which are determined by eccentric observations from the VLBI station. This paper shows KTRF94 coordinates of fundamental stations which accuracies are estimated 1 cm in horizontal and 3 cm in vertical. Also, the coordinates are compared to WGS84 and/or KGS95.