Academia.eduAcademia.edu

Figure 1 – uploaded by fatima es-sabery

See full PDFdownloadDownload figure
FIGURE 1. Global architecture of our combined classifier based on fuzzy Vader lexicon and a parallel deep belief networ!

Figure 1 Global architecture of our combined classifier based on fuzzy Vader lexicon and a parallel deep belief networ!

Related Figures (28)

TABLE 1. Polarity intensity.
TABLE 1. Polarity intensity.
FIGURE 2. Proportion of positive, neutral, and negative tweeted posts in the Sentiment140 corpus. F. Es-sabery et al/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data
FIGURE 2. Proportion of positive, neutral, and negative tweeted posts in the Sentiment140 corpus. F. Es-sabery et al/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data
FIGURE 4. Stages of Mamdani fuzzy system. Once both sentimental scores NS (Negativity score) = 0.425 and PS (Positivity score) = 0.575 have been computed. The implementation of Mamdani’s fuzzy logic, which is com- posed of three stages as illustrated in the figure 4, is the following stage.
FIGURE 4. Stages of Mamdani fuzzy system. Once both sentimental scores NS (Negativity score) = 0.425 and PS (Positivity score) = 0.575 have been computed. The implementation of Mamdani’s fuzzy logic, which is com- posed of three stages as illustrated in the figure 4, is the following stage.
TABLE 2. Lingual input and output metrics of Mamdani’s fuzzy system. Fuzzification phase: After the setting of lingual variables and lingual words, the subsequent stage is the implemen- tation of fuzzification operation [40] to the net rates of PS and NS, employing one of the membership functions (MFs) detailed by the equations (1), (2), (3), (4) and (5) for computing the membership degree u of both sentimental scores PS and NS in the Lower, Medium, and Higher fuzzy sets.
TABLE 2. Lingual input and output metrics of Mamdani’s fuzzy system. Fuzzification phase: After the setting of lingual variables and lingual words, the subsequent stage is the implemen- tation of fuzzification operation [40] to the net rates of PS and NS, employing one of the membership functions (MFs) detailed by the equations (1), (2), (3), (4) and (5) for computing the membership degree u of both sentimental scores PS and NS in the Lower, Medium, and Higher fuzzy sets.
A monotonically increasing membership function [42] is defined by two metrics d and p. It is defined by the next equation: A trapezoidal membership function [41] is defined by a low value lv, a high value hv and two values vp and vz which represent the boundaries of its kernel. The formula for the trapezoidal membership function is represented as follows:
A monotonically increasing membership function [42] is defined by two metrics d and p. It is defined by the next equation: A trapezoidal membership function [41] is defined by a low value lv, a high value hv and two values vp and vz which represent the boundaries of its kernel. The formula for the trapezoidal membership function is represented as follows:
Weighted Average Approach : is the simplest and, most commonly applied defuzzification method [50]. This tech- nique is also referred as the "Sugeno defuzzification" ap- proach. it can be formed by averaging every function of the outcome by its corresponding maximum belonging degree. Where x; denotes the element in the instance,u(a;) is the membership rate of the variable x; , and n presents the overall number of the variables in the used example. The representation graphic of the CM defuzzification approach is illustrated in the figure 6.
Weighted Average Approach : is the simplest and, most commonly applied defuzzification method [50]. This tech- nique is also referred as the "Sugeno defuzzification" ap- proach. it can be formed by averaging every function of the outcome by its corresponding maximum belonging degree. Where x; denotes the element in the instance,u(a;) is the membership rate of the variable x; , and n presents the overall number of the variables in the used example. The representation graphic of the CM defuzzification approach is illustrated in the figure 6.
Centroid approach : this method is also renowned as the centre of mass, of gravity, or of area [49]. It is the most frequently applied defuzzification technique. Its core idea is to identify the point x* at which a vertical boundary line would split the aggregate into two independent equal masses. It is defined by the following equation (11)
Centroid approach : this method is also renowned as the centre of mass, of gravity, or of area [49]. It is the most frequently applied defuzzification technique. Its core idea is to identify the point x* at which a vertical boundary line would split the aggregate into two independent equal masses. It is defined by the following equation (11)
Mean-Max Membership : this approach is also referred to as the medium of maxima procedure [51]. It is very closely linked to the maximum membership function, with the exception that the peak membership positions can be non- unique. The defuzzified outcome here is expressed by the subsequent equation (13):
Mean-Max Membership : this approach is also referred to as the medium of maxima procedure [51]. It is very closely linked to the maximum membership function, with the exception that the peak membership positions can be non- unique. The defuzzified outcome here is expressed by the subsequent equation (13):
has at least 2 non-overlapping fuzzy convex subsets. The out- come, in this scenario, is skewed to one side of a membership method [53]. Whenever the fuzzy outcome has two or more convex areas, so, the centroid of the convex fuzzy subarea with the biggest value is taken to get the defuzzified value x*. The value is determined by the next equation (15).
has at least 2 non-overlapping fuzzy convex subsets. The out- come, in this scenario, is skewed to one side of a membership method [53]. Whenever the fuzzy outcome has two or more convex areas, so, the centroid of the convex fuzzy subarea with the biggest value is taken to get the defuzzified value x*. The value is determined by the next equation (15).
vectors. In our contribution, we have applied four feature extractors, which are: GloVe, TFIDF (Trigram), TFIDF (Bi- gram), TFIDF (Unigram) in order to discover which one provides great accuracy rate.
vectors. In our contribution, we have applied four feature extractors, which are: GloVe, TFIDF (Trigram), TFIDF (Bi- gram), TFIDF (Unigram) in order to discover which one provides great accuracy rate.
FIGURE 13. Example of linear sub-structures when working with GloVe. Where X;; indicates the frequency with which word j occurs in the content of term i. 6s; and bs; are the biases related to the terms i and j respectively. E. STAGE OF FEATURE SELECTION
FIGURE 13. Example of linear sub-structures when working with GloVe. Where X;; indicates the frequency with which word j occurs in the content of term i. 6s; and bs; are the biases related to the terms i and j respectively. E. STAGE OF FEATURE SELECTION
TABLE 3. Defining the DBN model's parameters. Choosing the hyper-parameters, like the number of hidden units, RBM and DBN learning rates, the number of epochs, the batch size, the number of hidden and visible layers and the depth of features has a serious influence on the classification’s accuracy and the calculation’s complexity. The precision of DBN might not be more effective than
TABLE 3. Defining the DBN model's parameters. Choosing the hyper-parameters, like the number of hidden units, RBM and DBN learning rates, the number of epochs, the batch size, the number of hidden and visible layers and the depth of features has a serious influence on the classification’s accuracy and the calculation’s complexity. The precision of DBN might not be more effective than
MapReduce Programming Model
MapReduce Programming Model
FIGURE 15. data set before labeling. Figure 15 describes the datasets before the re-labeling process and the figure 16 shows the dataset after re-labeling. tion/defuzzification combination with the highest classifica- tion rate (98.96%) and lower error rate (1.04%).
FIGURE 15. data set before labeling. Figure 15 describes the datasets before the re-labeling process and the figure 16 shows the dataset after re-labeling. tion/defuzzification combination with the highest classifica- tion rate (98.96%) and lower error rate (1.04%).
From the figures 15 and 16, we remark that our hybrid process re-labeled the dataset into three class label: Negative label represents 34% , Positive label represent 44%, and Neu- tral represents 22% of the whole dataset instead of bot label Negative and Positive in the original dataset and h class every class label represents 50% of the whole dataset. According to this re-labelling process, we deduce that the original d is mislabelled with 22% as error rate. ata set
From the figures 15 and 16, we remark that our hybrid process re-labeled the dataset into three class label: Negative label represents 34% , Positive label represent 44%, and Neu- tral represents 22% of the whole dataset instead of bot label Negative and Positive in the original dataset and h class every class label represents 50% of the whole dataset. According to this re-labelling process, we deduce that the original d is mislabelled with 22% as error rate. ata set
FIGURE 18. Analysis of feature selectors. FIGURE 17. Evaluation of feature extractors. F. Es-sabery et a/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data
FIGURE 18. Analysis of feature selectors. FIGURE 17. Evaluation of feature extractors. F. Es-sabery et a/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data
TABLE 6. Classification rate, Error rate of different combinations Fuzzification/Defuzzification. TABLE 7. Classification rate, Error rate, Run-time and Kappa-Static of different combinations
TABLE 6. Classification rate, Error rate of different combinations Fuzzification/Defuzzification. TABLE 7. Classification rate, Error rate, Run-time and Kappa-Static of different combinations
TABLE 9. FalsePositiveRate and FalseNegativeRate of different combinations. the hybridization (TF-IDF Trigram+Hybridization+DBN) delivered the best performance in terms of classification rate (98.96%), error rate (1.04%), runtime (32.95s), kappa- static (99.10%), precision (98.96), recall(99.75%), Fl-score (99.53%), specificity (98.67%), false positive rate (0.25%) and false negative rate (1.33%). We were able to draw the conclusion from the compar- ative analysis that the outcomes produced by our approach significantly outperform those of the seven algorithms, This demonstrates the benefit of employing a deep learning tech- nique for tasks like these involving sentiment analysis and the strong results we obtain when dealing with large amounts of data as opposed to using conventional ML algorithms.
TABLE 9. FalsePositiveRate and FalseNegativeRate of different combinations. the hybridization (TF-IDF Trigram+Hybridization+DBN) delivered the best performance in terms of classification rate (98.96%), error rate (1.04%), runtime (32.95s), kappa- static (99.10%), precision (98.96), recall(99.75%), Fl-score (99.53%), specificity (98.67%), false positive rate (0.25%) and false negative rate (1.33%). We were able to draw the conclusion from the compar- ative analysis that the outcomes produced by our approach significantly outperform those of the seven algorithms, This demonstrates the benefit of employing a deep learning tech- nique for tasks like these involving sentiment analysis and the strong results we obtain when dealing with large amounts of data as opposed to using conventional ML algorithms.
TABLE 8. Precision, Recall, F1-Score and Specificity of different combinations.
TABLE 8. Precision, Recall, F1-Score and Specificity of different combinations.
F. Es-sabery et a/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data
F. Es-sabery et a/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data
TABLE 10. Complexity in term of space of our developed hybrid model, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65].
TABLE 10. Complexity in term of space of our developed hybrid model, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65].
F. Es-sabery et a/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data FIGURE 20. Comparative analysis of our method and deep learning algorithms.
F. Es-sabery et a/.: Emotion processing by applying a fuzzy-based Vader lexicon and a parallel deep belief network over massive data FIGURE 20. Comparative analysis of our method and deep learning algorithms.
FIGURE 21. Comparison of our model with other approaches selected from the literature.
FIGURE 21. Comparison of our model with other approaches selected from the literature.
TABLE 11. Complexity in terms of time of our developed hybrid model, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al [65]. In the fifth experiment, we have assessed the efficiency of our proposed hybrid method and the Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65] methods chosen from the literature to train the Sentiment140 dataset by demonstrating the convergence of each evaluated approach using the equation 52 in order to determine the iteration number when the analyzed method verified the condition described below in the following equa- tion 52.
TABLE 11. Complexity in terms of time of our developed hybrid model, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al [65]. In the fifth experiment, we have assessed the efficiency of our proposed hybrid method and the Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65] methods chosen from the literature to train the Sentiment140 dataset by demonstrating the convergence of each evaluated approach using the equation 52 in order to determine the iteration number when the analyzed method verified the condition described below in the following equa- tion 52.
TABLE 12. Convergence round of our innovative hybrid approach, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65]. In the final experiment, we measured the mean standard deviation (MSD) of each approach in comparison to the various 5 cross-validations of the given corpus to examine the effectiveness of our developed hybrid pattern, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65]. The primary objective of this experiment is to find the more stable approaches amongst all the evaluated approaches.
TABLE 12. Convergence round of our innovative hybrid approach, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65]. In the final experiment, we measured the mean standard deviation (MSD) of each approach in comparison to the various 5 cross-validations of the given corpus to examine the effectiveness of our developed hybrid pattern, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65]. The primary objective of this experiment is to find the more stable approaches amongst all the evaluated approaches.
FIGURE 22. The convergence rate of our proposed hybrid model when it was performed over the Sentiment140 dataset.
FIGURE 22. The convergence rate of our proposed hybrid model when it was performed over the Sentiment140 dataset.
TABLE 13. Stability of our innovative hybrid approach, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65] ir comparison to the various 5 cross-validations.
TABLE 13. Stability of our innovative hybrid approach, Botchway et al. [7], Es-Sabery et al. [10], Hua et al. [63], Hassan et al. [64] and Chen et al. [65] ir comparison to the various 5 cross-validations.
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved