Academia.eduAcademia.edu

Figure 1 – uploaded by Ehsan Hoseinzade

See full PDFdownloadDownload figure
layers. This happens through Eq 10. The last step is updating weights of the CNN using calculated gradient

Figure 1 layers. This happens through Eq 10. The last step is updating weights of the CNN using calculated gradient

Related Figures (13)

that fits the problem.
that fits the problem.
called the base predictor. 4.3. Transfer learning
called the base predictor. 4.3. Transfer learning
a graphical visualization of what we described just now. Figure 1: Graphical Visualization of 2D-CNNpred (Hoseinzade & Haratizadeh, 2019)
a graphical visualization of what we described just now. Figure 1: Graphical Visualization of 2D-CNNpred (Hoseinzade & Haratizadeh, 2019)
Table 3: Description of used algorithms
Table 3: Description of used algorithms
Table 4: F-measure of some of the stocks in S&P 500 index
Table 4: F-measure of some of the stocks in S&P 500 index
Figure 2: Performance of algorithms in having the best F-measure for predicting each of the 458 stocks
Figure 2: Performance of algorithms in having the best F-measure for predicting each of the 458 stocks
Table 5: F-measure of base predictor with different layers
Table 5: F-measure of base predictor with different layers
Table 6: F-measure of algorithms in prediction of U.S. indices
Table 6: F-measure of algorithms in prediction of U.S. indices
Table 7: F-measure of algorithms in prediction of world's famous indices
Table 7: F-measure of algorithms in prediction of world's famous indices
The list of features that were used in this research is represented in Table 8:
The list of features that were used in this research is represented in Table 8:
References \hmadi, E., Jasemi, M., Monplaisir, L., Nabavi, M. A.. Mahmoodi, A., & Jam, P. A.
References \hmadi, E., Jasemi, M., Monplaisir, L., Nabavi, M. A.. Mahmoodi, A., & Jam, P. A.
Related topics:
Machine Learning
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved