Search alternatives:
largest decrease » larger decrease (Expand Search), marked decrease (Expand Search)
values decrease » values increased (Expand Search)
learning data » learning dataset (Expand Search), learning a (Expand Search)
data decrease » rate decreased (Expand Search), a decrease (Expand Search), deaths decreased (Expand Search)
a largest » _ largest (Expand Search), a large (Expand Search), a latest (Expand Search)
i values » _ values (Expand Search)
largest decrease » larger decrease (Expand Search), marked decrease (Expand Search)
values decrease » values increased (Expand Search)
learning data » learning dataset (Expand Search), learning a (Expand Search)
data decrease » rate decreased (Expand Search), a decrease (Expand Search), deaths decreased (Expand Search)
a largest » _ largest (Expand Search), a large (Expand Search), a latest (Expand Search)
i values » _ values (Expand Search)
-
1
-
2
The MAE value of the model under raw data.
Published 2025“…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. Further integration of Spearman correlation analysis and PCA dimensionality reduction created multidimensional feature sets, revealing substantial accuracy improvements: The BiLSTM model achieved an 83.6% cumulative MAE reduction from 1.65 (raw data) to 0.27 (STL-PCA), while traditional models like Prophet showed an 82.2% MAE decrease after feature engineering optimization. …”
-
3
Three error values under raw data.
Published 2025“…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. Further integration of Spearman correlation analysis and PCA dimensionality reduction created multidimensional feature sets, revealing substantial accuracy improvements: The BiLSTM model achieved an 83.6% cumulative MAE reduction from 1.65 (raw data) to 0.27 (STL-PCA), while traditional models like Prophet showed an 82.2% MAE decrease after feature engineering optimization. …”
-
4
-
5
Comparison of environmental perception time results at different learning rates.
Published 2025Subjects: -
6
-
7
-
8
Yearly crude rates normalized by their 1999 value.
Published 2025“…<p>Values below 1 indicate crude rate decreases relative to their 1999 value, those above 1 indicate increases. …”
-
9
-
10
Moran’s I value, Z value and P value of CCD-GFEE.
Published 2025“…Within them, the CCD of Chengdu is the highest, Chongqing has achieved the largest stage leap. (4) The global Moran’s I consistently remained positive and exhibited a tendency of initially rising and subsequently falling, indicating that the spatial aggregation effect of CCD-GFEE first increased and then decreased. (5) The CCD-GFEE driving factors are examined using the spatial econometric model, and it has been observed that the impact of population size and government intervention on CCD-GFEE is negative, while the impact of industrial structure, technological progress and economic level on the coupling and coordination of CCD-GFEE is positive. …”
-
11
-
12
Distribution of <i>L</i>(<i>p</i>) with an absolute error when a single observation is randomly sampled at each time point, with the smallest observation (blue) and the largest obs...
Published 2024“…As the <i>L</i>(<i>p</i>) distribution associated with the larger observation tends to have higher overall values, the likelihood of it being ultimately selected decreases.…”
-
13
-
14
-
15
-
16
-
17
-
18
-
19
Estimation and prediction of the incidence <i>I</i>(<i>t</i>) in synthetic data.
Published 2024“…For each of these scenarios, we consider four testing and reporting models (see <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1012687#sec009" target="_blank">Methods</a> for details): constant reporting exponent <i>n</i> = −1 and testing that grows indefinitely with incidence (left column); constant reporting exponent <i>n</i> = −2 and testing that grows indefinitely with incidence (center-left column); variable reporting exponent <i>n</i> = −<i>α</i> log <i>C</i> and testing that grows indefinitely with incidence (center-right column); variable reporting exponent <i>n</i> = −<i>α</i> log <i>C</i> and testing that saturates at high incidence values (right column). …”
-
20