Academia.eduAcademia.edu

Figure 3 – uploaded by Merve Işık

See full PDFdownloadDownload figure
models are shown in Figure 4.1. so the nearby words are more weighted. The general architectures of the Word2Vec

Figure 4 models are shown in Figure 4.1. so the nearby words are more weighted. The general architectures of the Word2Vec

Related Figures (16)

Figure 5.2. Structure of proposed CNN-based approach (testing and real-time) Figure 5.2 shows the mechanism of testing and/or using the approach in real-
Figure 5.2. Structure of proposed CNN-based approach (testing and real-time) Figure 5.2 shows the mechanism of testing and/or using the approach in real-
6.1.2 Turkish Dataset Table 6. 1 Sample of the English dataset's Tweets (Figure Eight, 2015). Since we could not find any dataset in Turkish, we had to create the dataset
6.1.2 Turkish Dataset Table 6. 1 Sample of the English dataset's Tweets (Figure Eight, 2015). Since we could not find any dataset in Turkish, we had to create the dataset
Table 6. 2 Samples of the collected Turkish tweets.
Table 6. 2 Samples of the collected Turkish tweets.
Figure 6.2. The accuracy and F1 score of the studied algorithms using the third and fourth datasets.
Figure 6.2. The accuracy and F1 score of the studied algorithms using the third and fourth datasets.
Table 6. 6 The execution time of each ensemble system for English.
Table 6. 6 The execution time of each ensemble system for English.
Experiment 5: Performance of Ensemble Systems for Turkish Language Table 6. 8 Execution time for ML algorithms for Turkish without Grid Search
Experiment 5: Performance of Ensemble Systems for Turkish Language Table 6. 8 Execution time for ML algorithms for Turkish without Grid Search
Table 6. 9 The accuracy of the studied Ensemble systems using Turkish datasets.
Table 6. 9 The accuracy of the studied Ensemble systems using Turkish datasets.
6.3.2 Investigating the Performance of Developed CNN Model Table 6. 10 The execution time of each ensemble system for Turkish datasets. In this section, the experiments for investigating the performance of tk developed CNN model and its results have been analyzed in detail. The main aims of this experiment are to find the degree that performance can
6.3.2 Investigating the Performance of Developed CNN Model Table 6. 10 The execution time of each ensemble system for Turkish datasets. In this section, the experiments for investigating the performance of tk developed CNN model and its results have been analyzed in detail. The main aims of this experiment are to find the degree that performance can
Experiment 3: Investigating the Effect of the Input Window Size. The aim of this experiment is to investigate the effect of changing the number
Experiment 3: Investigating the Effect of the Input Window Size. The aim of this experiment is to investigate the effect of changing the number
Table 6. 13 The effect of changing the number of words in the processed text window. Experiment 4: Effect of Changing the Number of CNN’s Filters
Table 6. 13 The effect of changing the number of words in the processed text window. Experiment 4: Effect of Changing the Number of CNN’s Filters
Figure 6.3. The Accuracy, Precision, Recall, and Fl Score for the Developed System.
Figure 6.3. The Accuracy, Precision, Recall, and Fl Score for the Developed System.
Related topics:
Artificial IntelligenceTurkish LanguageText ClassificationText Mining and Information Retrieval
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved