• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.1
• /
• pp.125-134
• /
• 2021

With the rapid development of the film industry, the number of films is significantly increasing and movie recommendation system can help user to predict the preferences of users based on their past behavior or feedback. This paper proposes a movie recommendation system based on the latent factor model with the adjustment of mean and bias in rating. Singular value decomposition is used to decompose the rating matrix and stochastic gradient descent is used to optimize the parameters for least-square loss function. And root mean square error is used to evaluate the performance of the proposed system. We implement the proposed system with Surprise package. The simulation results shows that root mean square error is 0.671 and the proposed system has good performance compared to other papers.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.6 no.1
• /
• pp.3-12
• /
• 1988

The objective of this paper is to study the characteristics of combination method to correct both angle and distance errors simultaneously based on the least square adjustment methods. Changing the standard errors of distance and angle, the simulation errors of triangultion, trilateration, and combination result in some 39.8%, 33.9%, and 26.3% respectively. As the above, combination method shows more consistent accuracy than other methods. When considering the weight factor about error elements with independence, the diminishing rate of simulated average standard error represents a various change in each method. But considering them simultaneously, it shows a remarkable rate of diminishing 75.5%, 74.1%, and 69.2% in each method. And also, by growing the weight factor, accuracy of triangulation method is growing, whereas that of trilateration is diminishing. Therefore, determining the reasonable weight factors of distance and angle errors simultaneously in the analytical combination method, this method is expected to be one of more accurate and more effective methods for determining horizontal positions on the earth.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.701-706
• /
• 1998

A design method of rule -based fuzzy modeling is presented for the model identification of complex and nonlinear systems. The proposed rule-based fuzzy modeling implements system structure and parameter identification in the efficient form of " IF..., THEN,," statements. using the theories of optimization and linguistic fuzzy implication rules. The improved complex method, which is a powerful auto-tuning algorithm, is used for tuning of parameters of the premise membership functions in consideration of the overall structure of fuzzy rules. The optimized objective function, including the weighting factors, is auto-tuned for better performance of fuzzy model using training data and testing data. According to the adjustment of each weighting factor of training and testing data, we can construct the optimal fuzzy model from the objective function. The least square method is utilized for the identification of optimum consequence parameters. Gas furance and a sewage treatment proce s are used to evaluate the performance of the proposed rule-based fuzzy modeling.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.13 no.1
• /
• pp.13-20
• /
• 1995

Error analysis is important in horizontal positioning. In case of error analysis, standard error ellipse is generally used to establish the precision regions, but it will be replaced by 95% confidence ellipse. It is more effective than standard error ellipse in resection for measured procedures and establishment for criterias of relative error. Therefore, In this paper deals with analysis of application to 95% confidence ellipse in horizontal positioning. This study deals with error analysis of plane trilateration network which are various types of control point. also, this paper have studied for theory of error analysis in order to using least square adjustment. Thus, This paper's conclusion are as follows; 95% confidence ellipse could be used to establish of specification in korea, also, it is possible for us to predesign for optimum surveying network by 95% confidence ellipse.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.5
• /
• pp.573-581
• /
• 2021

The current study analyzed the impacts of fiscal decentralization (FD) on the economic stability of Pakistan. This study used time series data from 1981 to 2017. The collected data was first passed through the unit root analysis. ARDL estimation techniques were employed to scrutinize the data where long-run associations were tested through Wald F-statistics. The long-run estimates were extracted by applying Ordinary Least Square, and error correction mechanisms were employed to find the speed of adjustment for disequilibria between the long and the short run. Wald F-statistics confirmed the existence of long-run cointegration. Long-run elasticities suggested that fiscal decentralization because of limited institutional capabilities of provincial governments failed in bringing stability in the economy of Pakistan. Similarly, transparency issues and misspecification of projects hinder the outcome of investment to stabilize the economy. High service payments on debt cut the amount that can be used for skills improvements and destabilize the economy. High Population growth puts pressure on infrastructure and reduces production capacity, ultimately destabilizing the economy by increasing unemployment and inflation. Based on these findings, the government is suggested to improve the institutional capacity of lower governments for the desired outcome of power devolution.

• Journal of Animal Science and Technology
• /
• v.44 no.1
• /
• pp.1-12
• /
• 2002

The environmental effects were studied to estimate age and lactation stage adjustment coefficients in the primary linear traits of Holstein dairy cattle in Korea. Calving year month, classifier, age-month at classification and lactation stage were the environmental factors which significantly affected the most of linear traits at the level of 0.01. F values of Stature, strength, body depth, thurl with and rear leg side view were relatively higher in the effects of age-month, and dairy form, rear udder height, rear udder width, and udder cleft had relatively higher F values in the effects of lactation stage. Udder depth were affected highly by both age-month and lactation stage. Through the least square means of traits and the interpolation and smoothing obtained by the regression analysis of log and exponential transformed adjustment coefficients, age-month and lactation stage coefficients were estimated, and applied to real data to check the variation in the age-month and lactation stage effects. The estimated mean squares showed that the variation in all the linear traits significantly were decreased for the adjusted factors without the significant changes of variation in calving year month and classifier. That udder depth adjusted for both the age-month and lactation stage resulted in the decreases of variation in the both effects.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.6
• /
• pp.699-710
• /
• 2009

If a country has her geoid model, it could be determine accurate orthometric height because the geoid model could provide continuous equi-gravity potential surface. And it is possible to improve the coordinates accuracy of national control points through geodetic network adjustment considering geoidal heights. This study aims to find the best way to develop geoid model in Philippines which have similar topographic conditions as like Malaysia and Indonesia in Eastsouth asia. So, in this study, it is surveyed the general theories of geoid determination and development cases of geoid model in Asia and it is computed that the geoidal heights and gravity anomalies by spherical harmonic analysis using EGM2008, the latest earth geopotential model. The results show that first, the development of gravimetric geoid model based on airborne gravimetry is needed and second, about 200 GPS surveying data at national benchmark is needed. It is concluded that it is the most reasonable way to develop the hybrid geoid model through fitting geometric geoid by GPS/leveling data to gravimetric geoid. Also, it is proposed that four band spherical Fast fourier transformation(FFT) method for evaluation of Stokes integration and remove and restore technique using EGM2008 and SRTM for calculation of gravimetric geoid model and least square collocation algorithm for calculation of hybrid geoid model.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.26 no.5
• /
• pp.505-512
• /
• 2008

The 3D Hough transformation is the one of the most powerful and popular algorithm for extracting plane parameters from LiDAR data. However, there are some problems when extracting building roof plane using 3D Hough transformation. This paper explains possible problems and solution for extracting roof plane. The algorithm defines peak plane, exact plane, and LESS plane for extracting accurate plane parameters in the accumulator of the 3D Hough transformation. The peak plane is the plane which is represented by peak in the accumulator. The exact plane is the plane which is represented by the accumulator cell which is closest to the actual plane. The LESS plane can be calculated from all LiDAR points in the exact plane by using least-square adjustment. Test results show that proposed algorithm can extracts building roof plane very accurately.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.337-340
• /
• 2004

The exactly reproduced prosody of a TTS system is one of the key factors that affect the naturalness of synthesized speech. In general, rules about prosody had been gathered either from linguistic knowledge or by analyzing the prosodic information from natural speech. But these could not be perfect and some of them could be incorrect. So we proposed artificial neural network(ANN)s that can be trained to team the prosody of natural speech and generate it. In learning phase, let ANNs learn the pitch and energy contour of center phoneme by applying a string of phonemes in a sentence to ANNs and comparing the output pattern with target pattern and making adjustment in weighting values to get the least mean square error between them. In test phase, the estimation rates were computed. We saw that ANNs could generate the prosody of a sentence.

• Proceedings of the Korean Vacuum Society Conference
• /
• 1994.02a
• /
• pp.43-43
• /
• 1994

We studied chemical reactio of Eu metal on the in situ cleaved CdTe(110) surface by pphotoemission sppectroscoppy using synchrotron radiation. The chamber was maintained with base ppressure $\leq$2${\times}$10-10 mb during the expperiment. The expperiment was carried out in pphoton Factory in Jappan. Core level pphotoemission sppectroscoppy was carried out with Al K${\alpha}$ Line. The CdTe simiconductor was determined to be pp-typpe with low dopping concentration from Hall measurement. We found that there are two reacted pphases of Te with Eu (related to divalent Eu and trivalent Eu, resppectively) from least square fitting of Te 4d sppectra, but three is no indication of Cd reaction. Trivalent Eu exists after roughly one monolayer depposition (600 sec. depposition time is considered as one monolayer), which is also observed at Eu 3d core level sppectra. Overlayer Eu is metallized after roughly 2 monolayers depposition, as can be deduced from the fact that metallic edge near Fermi level begins to appear. The intensity of core-level of Te decreases expponentially at the initial stage (near one monolayer) and after one monolayer depposition it decreases more slowly due to Te out-diffusion. We categorized the growth mode of Eu on CdTe as S-K growth mode (cluster formation after one monolayer deppisition) from the relative intensity pplot of Te 4d normalized to the cleaved surface. At cleaved surface band bending is already established due to surface defects. At first 100 sec. depposition time the shift toward lower binding side by 0.6 eV is found at all core level sppectra of all elements in semiconductor. This shift is considered as the re-adjustment of surface Fermi level to the pposition induced by Eu metal (0.2 eV above the valence band maximum).