Raqamli texnologiyalarning Yangi O‘zbekiston rivojiga ta’siri


RAQAMLI TEXNOLOGIYALARNING



Yüklə 109,74 Kb.
Pdf görüntüsü
səhifə203/355
tarix27.12.2023
ölçüsü109,74 Kb.
#200050
1   ...   199   200   201   202   203   204   205   206   ...   355
15-8-PB

 
RAQAMLI TEXNOLOGIYALARNING 
YANGI 
O‘ZBEKISTON
 RIVOJIGA 
TA’SIRI
 
Xalqaro ilmiy-amaliy konferensiyasi
 
conducted in-depth research on three algorithm models, including the support vector machine 
algorithm model, artificial neural network algorithm model, and long and short-term memory 
network algorithm model. [Dahai Wang, (2022)] 
Claveria et al. introduced a sentiment construction method based on the evolution of survey-
based indicators using genetic algorithms. The study aimed to generate country-specific empirical 
economic sentiment indicators in the Baltic republics and the European Union. By searching for the 
optimal non-
linear combination of firms' and households’ expectations, the
authors computed the 
frequency distribution of survey expectations and examined the lag structure per selected variable. 
The study evaluated the out-of-sample predictive performance of the generated indicators, 
demonstrating more accurate estimates of year-on-year GDP growth rates compared to scaled 
industrial and consumer confidence indicators. The authors further combined the evolved 
expectations of firms and consumers using non-linear constrained optimization to generate 
aggregate expectations of year-on-year GDP growth, which outperformed recursive autoregressive 
predictions of economic growth. [Oscar CLAVERIA, (2021)] 
Overall, the reviewed studies emphasize the potential of machine learning algorithms, 
genetic algorithms, and emotion recognition algorithms in improving GDP prediction and economic 
forecasting accuracy. These innovative approaches offer new insights and methodologies for 
researchers and policymakers seeking more precise and reliable predictions of economic growth. 
By leveraging the power of advanced computational techniques, these studies contribute to the 
growing body of literature on machine learning-based forecasting methods, highlighting their 
superiority over traditional statistical models and the potential for enhancing decision-making in 
economic analysis and planning. 

Yüklə 109,74 Kb.

Dostları ilə paylaş:
1   ...   199   200   201   202   203   204   205   206   ...   355




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2024
rəhbərliyinə müraciət

gir | qeydiyyatdan keç
    Ana səhifə


yükləyin