• 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.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.173-173
• /
• 2023

The Mekong River Basin (MRB) is a crucial watershed in Asia, impacting over 60 million people across six developing nations. Accurate satellite-based precipitation products (SPPs) are essential for effective hydrological and watershed management in this region. However, the performance of SPPs has been varied and limited. The APHRODITE product, a unique gauge-based dataset for MRB, is widely used but is only available until 2015. In this study, we present a novel framework for correcting SPPs in the MRB by employing a deep learning approach that combines convolutional neural networks and encoder-decoder architecture to address pixel-by-pixel bias and enhance accuracy. The DLF was applied to four widely used SPPs (TRMM, CMORPH, CHIRPS, and PERSIANN-CDR) in MRB. For the original SPPs, the TRMM product outperformed the other SPPs. Results revealed that the DLF effectively bridged the spatial-temporal gap between the SPPs and the gauge-based dataset (APHRODITE). Among the four corrected products, ADJ-TRMM demonstrated the best performance, followed by ADJ-CDR, ADJ-CHIRPS, and ADJ-CMORPH. The DLF offered a robust and adaptable solution for bias correction in the MRB and beyond, capable of detecting intricate patterns and learning from data to make appropriate adjustments. With the discontinuation of the APHRODITE product, DLF represents a promising solution for generating a more current and reliable dataset for MRB research. This research showcased the potential of deep learning-based methods for improving the accuracy of SPPs, particularly in regions like the MRB, where gauge-based datasets are limited or discontinued.

• Journal of IKEEE
• /
• v.26 no.1
• /
• pp.77-82
• /
• 2022

The effect of the bias network on the performance of the W-band power amplifier(PA) was investigated. The performances of the typical RC and radial stub networks were examined, and a hybrid network was proposed for W-band application and its performance was confirmed. To verify this, a W-band PA was designed using a 60-nm GaN pHEMT process. When hybrid networks were applied, the PA had improved stability in all frequency bands, secured about 9 dB of power gain at operating frequencies 87 GHz to 93 GHz, and the maximum PAE was found to be about 12.3% at output power of 26.7 dBm.

• Korean Journal of Cognitive Science
• /
• v.28 no.3
• /
• pp.133-147
• /
• 2017

Human perception is an outcome of the influence of various factors rather than objective reflection of external environment. Among the factors, action-specific perception is a phenomenon where perception changes in terms of one's ability to act on the environment. Previous research reported contradictory results regarding whether action-specific perception occurs during performance or after performance due to memory distortion or knowledge about performance results. In this research, we conducted three experiments to determine when action-specific perception occurs. Participants threw darts at different distances and reported perceived size of targets in each trial. The results showed that, in Experiments 1 and 2, participants perceived targets larger when they hit than missed the targets, and the effect was greater when the targets were not visible after each throw. However, because participants had knowledge about the results of their throws, there could have been bias in participants' responses. In Experiment 3, where this possibility was excluded, we also obtained action-specific perception, and therefore concluded that action-specific perception occurs during but not after task performance.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.120-126
• /
• 2019

Recently, accurate prediction of power consumption based on machine learning techniques in Internet of Energy (IoE) has been actively studied using the large amount of electricity data acquired from advanced metering infrastructure (AMI). In this paper, we propose a deep learning model based on Gated Recurrent Unit (GRU) as an artificial intelligence (AI) network that can effectively perform pattern recognition of time series data such as the power consumption, and analyze performance of the prediction based on real household power usage data. In the performance analysis, performance comparison between the proposed GRU-based learning model and the conventional learning model of Long Short Term Memory (LSTM) is described. In the simulation results, mean squared error (MSE), mean absolute error (MAE), forecast skill score, normalized root mean square error (RMSE), and normalized mean bias error (NMBE) are used as performance evaluation indexes, and we confirm that the performance of the prediction of the proposed GRU-based learning model is greatly improved.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1489-1492
• /
• 1997

To increase reliability and performance of an IMS(Intelligent Manufacturing System), fault tolerant control based on an accurate fault diagnosis is needed. In this paper, robot FDI(fault detection and identification) is proposed for IMS where the robot is controlled with state estimates of a nonlinear filter using a mathematical robot model. The Chi-square distribution is applied fault detection and fault size is estimated by a proposed bias filter. Performance of the proposed algorithm is tested by simulation for studies.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.32.3-32
• /
• 2002

We analyze the effect of main inertial sensor errors such as, misalignment, scale factor error and bias on the performance of modified sequential probability ratio test (SPRT) for sequential fault detection and isolation (FDI). The inertial sensor errors cause the modified SPRT method to give false alarm. We use a two-stage KF to obtain a modified parity vector with which the inertial sensor errors can be removed and thus modified SPRT method can be used regardless of inertial sensor errors.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.5
• /
• pp.666-673
• /
• 1998

To increase reliability and performance of an IMS(Intelligent Manufacturing System), fault tolerant control based on an accurate fault diagnosis is needed. In this paper, robot FDI(fault detection and identification) is proposed for IMS where the robot is controlled with state estimates of a nonlinear filter using a mathematical robot model. The Chi-square test and GLR(General likelihood ratio) test are applied for fault detection and fault size is estimated by a proposed bias filter. Performance of the proposed algorithm is tested by simulation for studies.

• 전자공학회논문지 IE
• /
• v.44 no.2
• /
• pp.14-20
• /
• 2007

This paper presents quaternary D flip-flop with advanced performance. Quaternary D flip-flop is composed of the components such as thermometer code output circuit, EX-OR gate, bias inverter, transmission gate and binary D flip-flop circuit. The designed circuit is simulated by HSPICE in $0.35{\mu}m$ one-poly six-metal CMOS process parameters with a single +3.3V supply voltage. In the simulations, sampling frequencies is measured around 100MHz. The PDP parameters and FOM we estimated to be 59.3fJ, 33.7 respectively.

• Journal of Technology Innovation
• /
• v.19 no.1
• /
• pp.29-53
• /
• 2011

This paper analyzes the effects of government-sponsored R&D on firm's employment and management performance, using the panel data of manufacturing firms in the area of Busan, Ulsan and Gyungnam. The paper applies PSM to estimate the treatment effects(ATT) without sample selection bias. The findings show that government-sponsored R&D in the area has positive effects on the participating firms' employment and R&D for several years after completing R&D projects.