يعرض 3,121 - 3,140 نتائج من 21,342 نتيجة بحث عن '(( significance ((a decrease) OR (greatest decrease)) ) OR ( significant decrease decrease ))', وقت الاستعلام: 0.50s تنقيح النتائج
  1. 3121
  2. 3122
  3. 3123
  4. 3124

    Prime sequences. حسب Qiaozhen Qin (13159201)

    منشور في 2025
    الموضوعات:
  5. 3125
  6. 3126

    Cultural heritage dataset. حسب Jie Li (15030)

    منشور في 2025
    الموضوعات:
  7. 3127
  8. 3128
  9. 3129
  10. 3130
  11. 3131
  12. 3132
  13. 3133
  14. 3134
  15. 3135

    Achieving Improved Ion Swarm Shaping Based on Ion Leakage Control in Ion Mobility Spectrometry حسب Jiyao Wang (2121157)

    منشور في 2025
    "…Simulations and experiments demonstrate that precise voltage adjustments effectively minimize ion leakage, enhancing resolving power by 50% (reaching a maximum of 106), while the corresponding decrease in signal intensity follows the <i>I</i><sub>p</sub>–<i>R</i><sub>p</sub> linear relationship. …"
  16. 3136

    Testing set error. حسب Xiangjuan Liu (618000)

    منشور في 2025
    "…First, seven benchmark models including Prophet, ARIMA, and LSTM were applied to raw price series, where results demonstrated that deep learning models significantly outperformed traditional methods. Subsequently, STL decomposition decoupled the series into trend, seasonal, and residual components for component-specific modeling, achieving a 22.6% reduction in average MAE compared to raw data modeling. …"
  17. 3137

    Internal structure of an LSTM cell. حسب Xiangjuan Liu (618000)

    منشور في 2025
    "…First, seven benchmark models including Prophet, ARIMA, and LSTM were applied to raw price series, where results demonstrated that deep learning models significantly outperformed traditional methods. Subsequently, STL decomposition decoupled the series into trend, seasonal, and residual components for component-specific modeling, achieving a 22.6% reduction in average MAE compared to raw data modeling. …"
  18. 3138

    Prediction effect of each model after STL. حسب Xiangjuan Liu (618000)

    منشور في 2025
    "…First, seven benchmark models including Prophet, ARIMA, and LSTM were applied to raw price series, where results demonstrated that deep learning models significantly outperformed traditional methods. Subsequently, STL decomposition decoupled the series into trend, seasonal, and residual components for component-specific modeling, achieving a 22.6% reduction in average MAE compared to raw data modeling. …"
  19. 3139

    The kernel density plot for data of each feature. حسب Xiangjuan Liu (618000)

    منشور في 2025
    "…First, seven benchmark models including Prophet, ARIMA, and LSTM were applied to raw price series, where results demonstrated that deep learning models significantly outperformed traditional methods. Subsequently, STL decomposition decoupled the series into trend, seasonal, and residual components for component-specific modeling, achieving a 22.6% reduction in average MAE compared to raw data modeling. …"
  20. 3140

    Analysis of raw data prediction results. حسب Xiangjuan Liu (618000)

    منشور في 2025
    "…First, seven benchmark models including Prophet, ARIMA, and LSTM were applied to raw price series, where results demonstrated that deep learning models significantly outperformed traditional methods. Subsequently, STL decomposition decoupled the series into trend, seasonal, and residual components for component-specific modeling, achieving a 22.6% reduction in average MAE compared to raw data modeling. …"