• Annual Conference on Human and Language Technology
• /
• 2012.10a
• /
• pp.58-61
• /
• 2012

본 논문에서는 트위터 사용자들의 의견을 바이어스 탐지 하기 위해, 핵심 자질 추출 방법으로 HITS 그래프를 이용한 방법을 제안한다. 제안하는 핵심 자질 추출 방법은 사람이 직접 추출하지 못하는 자질도 추출할 수 있는 장점을 보였다. 제안한 핵심 자질 추출이 바이어스 탐지에 유효함을 검증하기 위해 4개의 토픽에 대해 평가 했을 때 제안 모델이 기존 모델보다 우수한 성능을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.398-401
• /
• 2012

사람들은 기업이나 제품에 대해 자신의 생각이 긍정적인지 부정적인지 표현하고자 한다. 트위터 사용자들은 트윗을 통해 자신의 생각을 표현한다. 본 논문에서는 트위터 데이터를 대상으로 대립관계에 있는 이슈에서의 바이어스 탐지 방법을 제안한다. 비지도학습 방법을 이용하여 트윗 패턴을 통해 세부자질을 추출하며, 세부자질에 대한 감정에 따른 확률 테이블을 구축하여 바이어스 탐지를 수행한다. 제안 방법의 유효성을 검증하기 위해 4 개의 대립 이슈에 대해 평가를 하였으며, 제안 모델이 기존의 모델보다 우수한 성능을 보였다.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.1-9
• /
• 2023

Stress is a significant issue in modern society, often triggered by external or internal factors that are difficult to manage. When high stress persists over a long term, it can develop into a chronic condition, negatively impacting health and overall well-being. However, it is challenging for individuals experiencing chronic stress to recognize their condition, making early detection and management crucial. Using biosignals measured from wearable devices to detect stress could lead to more effective management. However, there are two main problems with using biosignals: first, manually extracting features from these signals can introduce bias, and second, the performance of classification models can vary greatly depending on the subject of the experiment. This paper proposes a model that reduces bias using convo utional autoencoders, which can represent the key features of data, and enhances generalizability by employing soft voting, a method of ensemble learning, to minimize performance variability. To verify the generalization performance of the model, we evaluate it using LOSO cross-validation method. The model proposed in this paper has demonstrated superior accuracy compared to previous studies using the WESAD dataset.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.1
• /
• pp.22-28
• /
• 2021

Multivariate processes, such as chemical and mechanical process, power plants are operated in a state where several facilities are complexly connected, the fault of a particular system can also have fatal consequences for the entire process. In addition, since process data is measured in an unstable environment, outlier is likely to be include in the data. Therefore, monitoring technology is essential, which can remove outlier from measured data and detect failures in advance. In this paper, data obtained from dynamic and multivariate process models was used to detect fault in various type of processes. The dynamic process is a simulation of a process with autoregressive property, and the multivariate process is a model that describes a situation when a specific sensor fault. Mahalanobis distance was used to remove outlier contained in the data generated by dynamic process model and multivariate process model, and fault detection was performed using ICA. For comparison, we compared performance with and a conventional single ICA method. The proposed fault detection method improves performance by 0.84%p for bias data and 6.82%p for drift data in the dynamic process. In the case of the multivariate process, the performance was improves by 3.78%p, therefore, the proposed method showed better fault detection performance.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.4
• /
• pp.433-442
• /
• 2010

In this paper, self-calibration algorithm based on covariance matrix is proposed for compensating amplitude/phase mismatch in multi-baseline interferometer direction finding system. The proposed method is a solution to nonlinear constrained minimization problem which dramatically calibrate mismatch error using space sector concept with cost function as defined in this paper. This method, however, has a drawback that requires an estimated initial angle to determine the proper space sector. It is well known that this type of drawback is common in nonlinear optimization problem. Superior calibration capabilities achieved with this approach are illustrated by simulation experiments in comparison with interferometer algorithm for a varitiety of amplitude/phase mismatch error. Furthermore, this approach has been found to provide an exceptional calibration capabilities even in case amplitude and phase mismatch are more than 30 dB and over $5^{\circ}$, respectively, with sector spacing of less than $50^{\circ}$.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.33 no.6
• /
• pp.255-260
• /
• 2023

As demanding the detection of explosive molecules, it is required to develop rapidly and precisely responsive sensors with ultra-high sensitivity. Since two-dimensional semiconductors have an atomically thin body nature where mobile carriers accumulate, the abrupt modulation carrier in the thin body channel can be expected. To investigate the effectiveness of WSe₂ semiconductor materials as a detection material for TNT (Trinitrotoluene) explosives, WSe₂ was synthesized using thermal chemical vapor deposition, and afterward, WSe₂ FETs (Field Effect Transistors) were fabricated using standard photo-lithograph processes. Raman Spectrum and FT-IR (Fourier-transform infrared) spectroscopy reveal that the adsorption of TNT molecules induces the structural transition of WSe₂ crystalline. The electrical properties before and after adsorption of TNT molecules on the WSe₂ surface were compared; as -50 V was applied as the back gate bias, 0.02 μA was recorded in the bare state, and the drain current increased to 0.41 μA with a dropping 0.6% (w/v) TNT while maintaining the p-type behavior. Afterward, the electrical characteristics were additionally evaluated by comparing the carrier mobility, hysteresis, and on/off ratio. Consequently, the present report provides the milestone for developing ultra-sensitive sensors with rapid response and high precision.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.10
• /
• pp.758-765
• /
• 2018

This paper presents a 24-GHz frequency-modulated continuous wave(FMCW) radar transceiver with two Rx and one Tx channels in 65-nm complementary metal-oxide-semiconductor(CMOS) process and implemented it on a radar system using the developed transceiver chip. The transceiver chip includes a $14{\times}$ frequency multiplier, low-noise amplifier, down-conversion mixer, and power amplifier(PA). The transmitter achieves >10 dBm output power from 23.8 to 24.36 GHz and the phase noise is -97.3 GHz/Hz at a 1-MHz offset. The receiver achieves 25.2 dB conversion gain and output $P_{1dB}$ of -31.7 dBm. The transceiver consumes 295 mW of power and occupies an area of $1.63{\times}1.6mm^2$. The radar system is fabricated on a low-loss Duroid printed circuit board(PCB) stacked on the low-cost FR4 PCBs. The chip and antenna are placed on the Duroid PCB with interconnects and bias, gain blocks and FMCW signal-generating circuitry are mounted on the FR4 PCB. The transmit antenna is a $4{\times}4$ patch array with 14.76 dBi gain and receiving antennas are two $4{\times}2$ patch antennas with a gain of 11.77 dBi. The operation of the radar is evaluated and confirmed by detecting the range and azimuthal angle of the corner reflectors.