• 제목/요약/키워드: Performance Data

검색결과 31,715건 처리시간 0.049초

Improvement of Control Performance by Data Fusion of Sensors

  • Na, Seung-You;Shin, Dae-Jung
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • 제4권1호
    • /
    • pp.63-69
    • /
    • 2004
  • In this paper, we propose a general framework for sensor data fusion applied to control systems. Since many kinds of disturbances are introduced to a control system, it is necessary to rely on multisensor data fusion to improve control performance in spite of the disturbances. Multisensor data fusion for a control system is considered a sequence of making decisions for a combination of sensor data to make a proper control input in uncertain conditions of disturbance effects on sensors. The proposed method is applied to a typical control system of a flexible link system in which reduction of oscillation is obtained using a photo sensor at the tip of the link. But the control performance depends heavily on the environmental light conditions. To overcome the light disturbance difficulties, an accelerometer is used in addition to the existing photo sensor. Improvement of control performance is possible by utilizing multisensor data fusion for various output responses to show the feasibility of the proposed method in this paper.

음성/데이터 통합형 PRMA 프로토콜의 성능 개선 기법 (A Method for the Performance Ehancement of PRMA Protocol for Mobile Voice/Data Integration)

  • 송재섭;김연수
    • 한국통신학회논문지
    • /
    • 제25권3B호
    • /
    • pp.423-430
    • /
    • 2000
  • Future microcellular systems will require distributed network control. A packet-switched network is suitable for this requirement. The packet reservation multiple access(PRMA) is a Reservation-ALOHA like protocol for wireless terminals to transmit packet speech to a base station. It allows spatially distributed users in cellular systems to transmit packeted voice and data to a common base station using a shared channel. In the existing PRMA, the problem is that the voice packets may collide with the data packets due to simultaneous channel access. the problem may be a major performance degradation factor to a voice and data mixed system. We propose a new PRMA method that integrates voice and data traffic efficiently by resolving the collision problem between data and voice packets. The proposed PRMA method gives a performance improvement than the existing PRAMA method in terms of voice packet dropping probability and data delay characteristic. From analytic results, we can confirm that the proposed PRMA method show a performance improvement than the existing PRMA protocol.

  • PDF

A Study on Korean Sentiment Analysis Rate Using Neural Network and Ensemble Combination

  • Sim, YuJeong;Moon, Seok-Jae;Lee, Jong-Youg
    • International Journal of Advanced Culture Technology
    • /
    • 제9권4호
    • /
    • pp.268-273
    • /
    • 2021
  • In this paper, we propose a sentiment analysis model that improves performance on small-scale data. A sentiment analysis model for small-scale data is proposed and verified through experiments. To this end, we propose Bagging-Bi-GRU, which combines Bi-GRU, which learns GRU, which is a variant of LSTM (Long Short-Term Memory) with excellent performance on sequential data, in both directions and the bagging technique, which is one of the ensembles learning methods. In order to verify the performance of the proposed model, it is applied to small-scale data and large-scale data. And by comparing and analyzing it with the existing machine learning algorithm, Bi-GRU, it shows that the performance of the proposed model is improved not only for small data but also for large data.

Evaluating Chest Abnormalities Detection: YOLOv7 and Detection Transformer with CycleGAN Data Augmentation

  • Yoshua Kaleb Purwanto;Suk-Ho Lee;Dae-Ki Kang
    • International journal of advanced smart convergence
    • /
    • 제13권2호
    • /
    • pp.195-204
    • /
    • 2024
  • In this paper, we investigate the comparative performance of two leading object detection architectures, YOLOv7 and Detection Transformer (DETR), across varying levels of data augmentation using CycleGAN. Our experiments focus on chest scan images within the context of biomedical informatics, specifically targeting the detection of abnormalities. The study reveals that YOLOv7 consistently outperforms DETR across all levels of augmented data, maintaining better performance even with 75% augmented data. Additionally, YOLOv7 demonstrates significantly faster convergence, requiring approximately 30 epochs compared to DETR's 300 epochs. These findings underscore the superiority of YOLOv7 for object detection tasks, especially in scenarios with limited data and when rapid convergence is essential. Our results provide valuable insights for researchers and practitioners in the field of computer vision, highlighting the effectiveness of YOLOv7 and the importance of data augmentation in improving model performance and efficiency.

Performance Analysis of Transmit Diversity in Multiuser Data Networks With Fading Correlation

  • Zhang, Kai;Niu, Zhisheng
    • Journal of Communications and Networks
    • /
    • 제10권4호
    • /
    • pp.444-450
    • /
    • 2008
  • This paper studies the performance of multiuser data networks with transmit diversity under correlated fading channels. Previous work shows that correlated fading reduces the link performance of multiple antenna systems, but how correlated fading affects the throughput of multiuser data networks is still unknown since the throughput depends not only on the link performance but also on the multiuser diversity. We derive the throughput of the multiuser data networks with various transmit diversity schemes under correlated fading channels. The impact of correlated fading on the throughput is investigated. Analytical and simulation results show that, although correlated fading is harmful for link performance, it increases the throughput of the multiuser data networks if the transmit scheme is appropriately selected.

Study Factors for Student Performance Applying Data Mining Regression Model Approach

  • Khan, Shakir
    • International Journal of Computer Science & Network Security
    • /
    • 제21권2호
    • /
    • pp.188-192
    • /
    • 2021
  • In this paper, we apply data mining techniques and machine learning algorithms using R software, which is used to predict, here we applied a regression model to test some factor on the dataset for which we assumed that it effects student performance. Model was built on an existing dataset which contains many factors and the final grades. The factors tested are the attention to higher education, absences, study time, parent's education level, parent's jobs, and the number of failures in the past. The result shows that only study time and absences can affect the students' performance. Prediction of student academic performance helps instructors develop a good understanding of how well or how poorly the students in their classes will perform, so instructors can take proactive measures to improve student learning. This paper also focuses on how the prediction algorithm can be used to identify the most important attributes in a student's data.

Study on the relationship between trust and organizational performance in local administrative organization- Focused on the local administrative organizations in Gangwondo-

  • Kim, Sun-Ok;Park, Sung-Yong;Lee, Hee-Choon
    • Journal of the Korean Data and Information Science Society
    • /
    • 제21권5호
    • /
    • pp.983-997
    • /
    • 2010
  • This study is to explore the relationship between trust in local administrative organization and organizational performance. Local administrative organizations provide the citizens with administrative services. Heightening the organizational performance contributes the citizens' happiness and the stream of times through organizations' change. To provide high quality of administrative service to citizens, trust in organizations is more important than any other capital. The improvement of organizational performance needs through this social capital. Factors about trust variables and organizational performance variables are extracted through the theoretical discussions. To do the research, public servants in 7 local administrative organizations of Gangwondo were asked to do the survey about how trust in organizations affects organizationa performances. The results explain that trust variables are related to organizational performance, and the local administrative organization which is high in trust is high in organizational performance. Trust in local administrative organizations improves the organizational performance internally and the organization will obtain trust from the citizens externally.

반사실적 데이터 증강에 기반한 인과추천모델: CausRec사례 (A Causal Recommendation Model based on the Counterfactual Data Augmentation: Case of CausRec)

  • 송희석
    • Journal of Information Technology Applications and Management
    • /
    • 제30권4호
    • /
    • pp.29-38
    • /
    • 2023
  • A single-learner model which integrates the user's positive and negative perceptions is proposed by augmenting counterfactual data to the interaction data between users and items, which are mainly used in collaborative filtering in this study. The proposed CausRec showed superior performance compared to the existing NCF model in terms of F1 value and AUC in experiments using three published datasets: MovieLens 100K, Amazon Gift Card, and Amazon Magazine. Compared to the existing NCF model, the F1 and AUC values of CausRec showed 1.2% and 2.6% performance improvement in MovieLens 100K data, and 2.2% and 10% improvement in Amazon Gift Card data, respectively. In particular, in experiments using Amazon Magazine data, F1 and AUC values were improved by 11.7% and 21.9%, respectively, showing a significant performance improvement effect. The performance of CausRec is improved because both positive and negative perceptions of the item were reflected in the recommendation at the same time. It is judged that the proposed method was able to improve the performance of the collaborative filtering because it can simultaneously alleviate the sparsity and imbalance problems of the interaction data.

Development of Performance Demonstration Programs for Eddy Current Data Analysis

  • Cho, Chan-Hee;Nam, Min-Woo;Yang, Seung-Han;Yang, Dong-Soon;Lee, Hee-Jong
    • 비파괴검사학회지
    • /
    • 제25권3호
    • /
    • pp.228-232
    • /
    • 2005
  • The Korea Electric Power Research Institute (KEPRI) has developed performance demonstration programs for non-destructive testing personnel who analyze ECT(eddy current testing) data for steam generator tubing since 2001 The purpose of these performance demonstration programs is to ensure a uniform knowledge and skill level of data analysts and contribute to safe operation of nuclear power plants. Many changes have occurred in non-destructive testing of steam generator tubing such as inspection scope, plugging criteria and qualification requirements. According to the Notice 2004-13 revised by the Ministry of Science and Technology (MOST), the analyst for steam generator tubing shall be qualified as the qualified data analyst (QDA), and the site specific performance demonstration (SSPD) program shall be implemented. KEPRI developed these performance demonstration programs and they are being successfully implemented. The analyst's performance is expected to be improved by the implementation of these programs.

Stream Processing에서 I/O데이터 일관성을 고려한 성능 최적화 (Performance Optimization Considering I/O Data Coherency in Stream Processing)

  • 나하나;이준환
    • 전자공학회논문지
    • /
    • 제53권8호
    • /
    • pp.59-65
    • /
    • 2016
  • 본 논문은 대량의 stream data를 처리하는 어플리케이션에서 하드웨어 가속기들이 접근하는 메모리가 non-cacheable에서 cacheable으로 변경됨에 따라 발생할 수 있는 데이터 일관성 문제를 고려하여 시스템 최적화를 진행하였다. 이를 위해 상위 수준 시뮬레이션을 통한 프로파일링 결과를 토대로 분석식을 만들어 활용하였다. 실험한 결과 여러 이미지 크기에서 메모리가 cacheable로 변경됨에 따라 평균 1.40배의 성능 향상을 보였다. 분석식의 주요 파라미터 최적화를 통해 최종적으로 3.88배의 성능 이득이 발생했으며, 항상 메모리가 cacheable인 경우의 성능이 항상 우월한 것은 아님을 확인할 수 있었다.