Emotions are a critical aspect of human behavior. One widely used technique for research in emotion measurement is based on the use of EEG signals. In general terms, the first step of signal processing is the elimination of noise, which can be done in manual or automatic terms. The next step is determining the feature vector using, for example, entropy calculation and its variations to generate a classification model. It is possible to use this approach to classify theoretical models such as the Circumplex model. This model proposes that emotions are distributed in a two-dimensional circular space. However, methods to determine the feature vector are highly susceptible to noise that may exist in the signal. In this article, a new method to adjust the classifier is proposed using metaheuristics based on the black hole algorithm. The method is aimed at obtaining results similar to those obtained with manual noise elimination methods. In order to evaluate the proposed method, the MAHNOB HCI Tagging Database was used. Results show that using the black hole algorithm to optimize the feature vector of the Support Vector Machine we obtained an accuracy of 92.56% over 30 executions.
Emotions play an important role regarding the way in which people think and behave [
In particular, the increase of the visual component P1 has been studied, with event-related potentials (ERP), by filtering low spatial frequencies, thus evidencing the rapid activation of the magnocellular system against stimuli that trigger emotions of high agitation [
One of the metrics (features) that is the most representative and which provides the most information is entropy. Entropy is a measurement of information or order; it measures the predictability of data. This is, given a set of
However, this use of entropy can magnify the signal noise, being extremely sensitive to minimal variations. For this reason, different ways of measuring entropy have been proposed, such as approximate entropy, differential entropy, or sample entropy. Among these methods, sample entropy presents a valuable statistical consistency and for this reason was utilized as a basis of comparison [
Although the SampEn method is highly accurate, it is extremely sensitive to its input parameters. In fact, there is no established consensus on the selection of parameters for small data sets, especially for biological data [
These situations present the problem of finding or calculating the most suitable value for entropy that allows generating high performance classifiers. This task is complex and can be seen as an optimization problem in itself. A first approximation to a potential solution can be the use of full-search algorithms to explore a tree of extremely large potential solutions. However, these techniques are highly costly and can lead to an unsuitable large amount of attempts to find a solution. With this is mind, it is not possible to propose complete techniques such as Backtracking or hybrid ones such as Forward Checking.
On the other hand, recently, several approaches have emerged, inspired by natural phenomena, that allow solving complex optimization and combinatorial problems in reduced time periods [
In this article, we propose using an approximate optimization approach to find the best values considering the predictability of the classifier. The reason for the proposed approach is the strong impact on the development of classifiers for emotion recognition based on electroencephalography. The main idea is to use the black hole algorithm due to its low cost, similar to the calculation of entropy. This algorithm is inspired by the phenomena of black holes [
The present work is organized as follows: Theoretical background is introduced in Section
First, in Section
To conclude, our proposal consists of the preprocessing of the signal (through EMD and sample entropy) for the construction of an initial multiclass SVM classifier. Using this classifier as a base, a population (group) of classifiers is created, which are formed by groups of modified characteristics coming from the initial characteristics and a random variation relative to the error of the classifier.
Once this population is created, it is iterated through the black hole metaheuristic, which continuously generates and improves these characteristics in order to obtain distinct classifiers; these classifiers are then evaluated, always, using the original characteristics from the signal. Once all the iterations are completed, the best classifier (historically speaking) is chosen; this classifier is, finally, utilized. Figure
Proposed approach.
Circumplex is one of the most used models for emotion classification [
Russell’s classification model [
There are variants of the Circumplex model in which extra dimensions are added, such as domination or freedom in a given situation [
In this work, we have used a discrete quadrant division to represent the greatest variation among emotional states. This approach is optimal for classifying and obtaining fewer error rates. This is because they represent the greatest possible distance between agitation and valence (the digital axes). On the other hand, it would be possible to classify discrete emotions; there would be a greater probability of erroneously classifying nearby emotions in this model because they would represent lower variance values.
Electroencephalography is a method of neurophysiological exploration that is based on the registry of cerebral activity through sensors that translate bioelectric activity into electrical current [
EEG signals are usually classified by their frequency, amplitude, shape, or electrode position. The EEG bands are
Even so, the position of sensors is standardized by the 10-20 channel system, by which each position is described by a combination of a letter and a number. The letter indicates the brain region that may be represented as frontal (F), central (C), temporal (T), occipital (O), or parietal (P) [
EEG 10-20 system.
There is also another positioning system named 10-10 system in which only the 10% proportion is used. In this alternative system the same bands mentioned before are used with the addition of other intermediate channels. In the case of the lobes, letter combinations are created for the channels between two regions, for example, FP for frontoparietal [
The assembly of the electrodes can be done by referencing the electrodes or with a bipole method. The reference is made with electrodes that generate a comparison link, generally with an electrode positioned in A2 (the ear electrode) and the bipole method is performed by recording the potential differences between paired electrodes [
The applications of EEG are varied [
Sample entropy is a variation of approximate entropy (ApEn). This entropy reduces the potential bias generated by self-matching that arises during ApEn [
Support Vector Machines (SVM) are a set of supervised learning algorithms based on statistics learning theory [
The main advantage of using SVMs is that their model can be generalized for nonlinear feature spaces. On the other hand, weighted SVM, which is the method used in this work, has a regularization parameter C that enables accommodation to outliers and allows errors on the training set.
Empirical Mode Decomposition (EMD) is a data-driven signal processing and analysis technique [
The main advantage of using this technique is that it permits softening the signals and decreasing noise, which is especially useful in physiological signals.
Each component fulfills 2 fundamental requirements: The number of endpoints and the number of crosses by zero (zero-crossings) is equal or differs at the most in 1. The average between the top and bottom wrapper is always zero at each point.
EMD generates a set of Intrinsic Mode Functions (IMF) that allows obtaining the components of a signal with most significance. The steps to define the set of functions are as follows: Identify all of the local endpoints of the signal. Connect all local maximums using cubic spline interpolation to create a superior wrapper. Repeat the same process for the local minimums. Create a The first resulting signal is the original signal minus The remainder of the original signal is obtained minus the IMF; i.e., If IMF satisfies the definition (the 2 basic requirements), it is accepted as a valid IMF; otherwise the process is rejected and repeated using the remainder as the original signal.
This continues until the stopping condition is met, which can be a certain number of iterations or until the residue contains no more than one endpoint.
The use of wavelet transformation for EEG signal classification was proposed by [
In spite of the mentioned limitations, the frequency analysis of wavelet has been used, among other things, to determine the intracortical coupling, unraveling cerebral synchrony through the systems of communication between near and distant neurons associated with cognitive processes [
EMD is an iterative process that allows a transversal time-frequency analysis by extracting the oscillatory characteristics. On the other hand, the wavelet transform allows performing a longitudinal analysis of the frequency changes over time by convolving a signal based on a mother wavelet. Particularly the EEG signals are characterized by being non-Gaussian and nonstationary; due to this, it has been observed that the wavelet transform has a worst resolution of time and frequency while the EMD provides a more intuitive understanding of the data [
WEAVE is EEG-emotion valence classifier based on five steps: Segmentation of EEG signals related to emotions in windows of 6 seconds. Extraction of the wavelet metrics to form WEAVE. Calculating the complexity of metrics with Normalized Mutual Information (NMI) [ Reduction of channels through NMI. Classification with the Support Vector Machine (SVM) algorithm using the Sequential Minimal Optimization (SMO) algorithm to train the SVM.
The advantages of the wavelet transform are due to the regularity in the intersegment estimation and the subbands obtainment through the bandpass filter and the denoiser signal decomposition [
A state of excitement in the cerebral cortex can be identified using the detection of a significant Beta Band [
In [
For the reconstruction of the Beta Band they used low pass and high pass Butterworth filters. Signals were filtered using a 3rd-order bandpass Butterworth filter [
Furthermore, for the experiment, the Database for Emotion Analysis using Physiological Signals (DEAP) was used [
EEG-based emotion recognition using combined feature extraction method.
Upon analyzing, in detail, the process, we can see that the entropy values strongly affect the creation of the classifier and are directly related to the configuration of the input parameters. In addition, due to the search process is an iterative procedure, it is not possible to determine the performance of the classifier until the process is finished.
For our proposal, presented in Section
Furthermore, for each session, participants were asked to answer a survey regarding emotions they felt, levels of agitation, valence and domination, among other questions. We used the agitation and valence (high, low) to create the multilabelled classifier, where each of the four classes is one of the quadrants. When using a multiclass model for classification, the answer must be in one of the classes contained in the model. To avoid the creation of a null class, it is advisable to use the full spectrum of emotions. For this, the Russell quadrant model was selected [
For this study, we used the F3 and C4 channels of the EEG sensor, as it was done in [
In optimization new approximate techniques have been proposed in order to improve the search process. Many of these algorithms are on inspired in social environments, natural phenomena, and the biological evolution [
To solve this problem, we propose to use an approximate method that permits evaluating previous behavior of the classifier, and if necessary, allowing for improvement. The approximate techniques have been widely used in real world problems [
The black hole algorithm is based on the phenomenon of the same name, which occurs in outer space and is inspired by the law of attraction/absorption. The algorithm follows three main fundamentals: A star in space is considered a solution to the problem. As a population-based algorithm, a certain number of stars are randomly generated. The black hole is selected. A black hole represents the star with the best performance of all solutions. The movement and generation of new stars are carried out through the absorption formula:
where
The event horizon is a radius originated by the black hole. In case a star crosses the horizon, it will be absorbed and destroyed by the black hole and a new star (solution) is created randomly. This is known as the probability of crossing the event horizon and is calculated as follows:
One of the most interesting characteristics of incomplete data processing algorithms is the approximation to good solutions. This concept may be used as stop criteria. However, in situations where the optimal solutions are not known a priori, it is not possible to measure the quality of found solutions. In these cases, possible stop criteria are the number of executed iterations, for the sake of clarity of the proposed algorithm. In our proposal, the stop criteria are initially set as 100 off-line iterations.
Algorithm
Randomness allows a degree of variability in the algorithm. Then, in the loop statement, the process of absorption of the algorithm is carried out. The quality of each solution is calculated, determined by the performance exhibited by the classifier. If the rating value is close to 1, the solution is considered to have a high quality (see Line
To measure the performance (quality) of the solutions, a proportion given by (
Finally, the loop statement ends when an adequate enough solution is reached for our approach; this condition is determined by updating the solution in a certain amount of iterations. At the end, the best solutions are memorized and visualized.
Figure
Proposed method using black hole algorithm.
After applying the approximation approach, we have analyzed the time complexity of the black hole algorithm into the process of creating the classifier and we illustrate that our proposal does not affect its performance. It can be determined that time complexity of the SampEn is given by
The performance of the black hole algorithm was experimentally evaluated by using a set of well-known validated signals using MAHNOB HCI Tagging Database [
The approximate approach has been implemented on the programming language C# and the experiments [
Parameter setting to the entropy and the black hole algorithm.
| | | |
---|---|---|---|
Data selection | Number of sessions | Emotion elicitation trials | 563 |
Frequency | Each second has 128 samples or values | 128 Hz | |
Frame | To classify each frame it lasts 9 seconds, without overlapping | 9 sec. | |
| |||
Sample entropy | | Number of samples | 128 samples |
| Embedded dimension | 2 | |
| Probability of similarity on two simultaneous datasets | 0.15 | |
| |||
Empirical Mode Decomposition | Order | Number of IMFs | 4 |
| |||
Black Hole Algorithm | | Number of stars (solutions) | 30 |
| Maximum iterations | 100 | |
| |||
Miscellaneous | – | Runs of the approximate approach | 30 |
– | Number of used cores (processors) | 8 |
Firstly, these parameter settings are adopted after a hard initial training phase, being the one that obtained the best results. Then, we considered previous works to compare the choice of parameter values as reported in [
A common method to recognize the emotion based on EEG signals uses the entropy factor to build the classifier. We have implemented this technique and the accuracy obtained was close to 84.77% producing an error of classification outperform to 15%. That can be attributed to the sample entropy that builds the classifier without iterating in order to find the best solution.
Towards the end of iterations, the approximate optimization method reaches an accuracy above to 93% illustrating again that its performance is better than sample entropy approach. All results are available in Appendix
Figure
Convergence chart of the proposed method.
It is possible to conclude that the results are promising compared to those obtained with other SVM classifiers built by using the entropy factor. The proposed method used the MAHNOB HCI Tagging Database and reached a maximum accuracy level of 93.03%, with an average of 92.57%. Using the same dataset, a standard approach using the entropy factor to build a SVM classifier presents an average accuracy of 84.77%. More details can be seen in Appendix
This approach could be useful in emotion classification if the research goal would be to obtain relevant information in real time, for instance, incorporating an EEG in the classroom [
Emotions have been subject to scientific research for more than a century, as they play many essential roles in people’s lives [
EEG-emotion signals allow for the prediction and classification of data with automated noise reduction. The emotion research is especially complex due to the ecological paradigm requirement, specifically the trigger stimuli, and emotional response generates high rate of noise. A common method is detailed in the background section, using entropy as a more relevant element. Nevertheless, results are not what was expected, reaching 85% in accuracy only.
In order to improve these computational results, we conducted an approximate method inspired on the black hole phenomenon. This algorithm is proposed to analyze the performance of an SVM classifier, allowing the extension of emotion ecological paradigms with EEG data.
We have tested our technique using a validated emotion signal, named MAHNOB HCI Tagging Database. Results show that the optimization algorithm allows the SMV classifier to surpass 90% in accuracy in its first iterations, even reaching 93%; furthermore, it is highly competitive with those presented in the related works section.
Particularly, these results are compatible with those obtained with the EEG-emotion signal with wavelet entropy and Support Vector Machine classifier proposed by Çelikkanat, but with higher accuracy [
As future works, we believe that using new approximate optimization algorithms will allow us to find better results to compare the SVM classifier performance. Moreover, we intend to incorporate an autonomous version of these algorithms so that the self-adaptive of its parameters is not complex and suited to the instance of the problem, as described in [
On the other hand, we propose an integration of autonomous search in the parameter settings process, in order to find the best values during the run. This research can lead towards new study lines.
In Tables
Computational results of the approximate approach.
| | |||
---|---|---|---|---|
Minimum | Average | Standard Deviation | Maximum | |
| 85.34 | 87.45 | 1.18E+04 | 89.33 |
| ||||
| 86.20 | 88.05 | 8.89E+03 | 89.61 |
| ||||
| 87.43 | 88.84 | 6.67E+03 | 90.28 |
| ||||
| 87.95 | 89.14 | 7.32E+03 | 91.18 |
| ||||
| 87.95 | 89.61 | 1.10E+04 | 92.13 |
| ||||
| 88.19 | 89.95 | 1.15E+04 | 92.13 |
| ||||
| 88.47 | 90.26 | 1.13E+04 | 92.27 |
| ||||
| 88.66 | 90.53 | 1.15E+04 | 92.46 |
| ||||
| 89.09 | 90.81 | 1.05E+04 | 92.46 |
| ||||
| 89.23 | 91.08 | 9.90E+03 | 92.50 |
| ||||
| 89.52 | 91.23 | 9.35E+03 | 92.50 |
| ||||
| 89.52 | 91.45 | 7.81E+03 | 92.55 |
| ||||
| 89.56 | 91.53 | 7.74E+03 | 92.55 |
| ||||
| 90.09 | 91.65 | 6.72E+03 | 92.55 |
| ||||
| 90.09 | 91.70 | 6.37E+03 | 92.55 |
| ||||
| 90.32 | 91.75 | 5.98E+03 | 92.60 |
| ||||
| 90.42 | 91.80 | 5.61E+03 | 92.60 |
| ||||
| 90.42 | 91.83 | 5.58E+03 | 92.60 |
| ||||
| 90.65 | 91.87 | 5.43E+03 | 92.60 |
| ||||
| 90.80 | 91.89 | 5.28E+03 | 92.60 |
Computational results of the approximate approach (continuation).
| | |||
---|---|---|---|---|
Minimum | Average | Standard Deviation | Maximum | |
| 90.99 | 91.95 | 4.94E+03 | 92.60 |
| ||||
| 91.03 | 91.98 | 4.64E+03 | 92.60 |
| ||||
| 91.13 | 92.01 | 4.48E+03 | 92.60 |
| ||||
| 91.13 | 92.04 | 4.37E+03 | 92.79 |
| ||||
| 91.13 | 92.05 | 4.41E+03 | 92.79 |
| ||||
| 91.13 | 92.08 | 4.35E+03 | 92.79 |
| ||||
| 91.13 | 92.09 | 4.33E+03 | 92.79 |
| ||||
| 91.13 | 92.10 | 4.27E+03 | 92.79 |
| ||||
| 91.13 | 92.12 | 4.27E+03 | 92.84 |
| ||||
| 91.13 | 92.14 | 4.17E+03 | 92.84 |
| ||||
| 91.13 | 92.16 | 4.15E+03 | 92.93 |
| ||||
| 91.18 | 92.17 | 4.13E+03 | 92.93 |
| ||||
| 91.22 | 92.18 | 4.08E+03 | 92.93 |
| ||||
| 91.27 | 92.19 | 4.12E+03 | 92.93 |
| ||||
| 91.32 | 92.21 | 3.99E+03 | 92.93 |
| ||||
| 91.32 | 92.23 | 3.91E+03 | 92.93 |
| ||||
| 91.56 | 92.25 | 3.67E+03 | 92.93 |
| ||||
| 91.56 | 92.26 | 3.53E+03 | 92.93 |
| ||||
| 91.56 | 92.28 | 3.53E+03 | 92.98 |
| ||||
| 91.56 | 92.29 | 3.52E+03 | 92.98 |
Computational results of the approximate approach (continuation).
| | |||
---|---|---|---|---|
Minimum | Average | Standard Deviation | Maximum | |
| 91.56 | 92.31 | 3.40E+03 | 92.98 |
| ||||
| 91.56 | 92.32 | 3.29E+03 | 92.98 |
| ||||
| 91.56 | 92.34 | 3.29E+03 | 92.98 |
| ||||
| 91.56 | 92.34 | 3.27E+03 | 92.98 |
| ||||
| 91.65 | 92.36 | 3.11E+03 | 92.98 |
| ||||
| 91.84 | 92.39 | 2.78E+03 | 92.98 |
| ||||
| 91.84 | 92.40 | 2.82E+03 | 92.98 |
| ||||
| 91.84 | 92.40 | 2.81E+03 | 92.98 |
| ||||
| 91.84 | 92.41 | 2.82E+03 | 92.98 |
| ||||
| 91.84 | 92.42 | 2.84E+03 | 92.98 |
| ||||
| 91.84 | 92.42 | 2.88E+03 | 93.03 |
| ||||
| 91.94 | 92.44 | 2.72E+03 | 93.03 |
| ||||
| 91.94 | 92.44 | 2.75E+03 | 93.03 |
| ||||
| 91.94 | 92.44 | 2.73E+03 | 93.03 |
| ||||
| 91.94 | 92.45 | 2.76E+03 | 93.03 |
| ||||
| 91.94 | 92.46 | 2.84E+03 | 93.03 |
| ||||
| 91.94 | 92.47 | 2.84E+03 | 93.03 |
| ||||
| 91.94 | 92.47 | 2.79E+03 | 93.03 |
| ||||
| 91.94 | 92.48 | 2.80E+03 | 93.03 |
| ||||
| 91.94 | 92.49 | 2.74E+03 | 93.03 |
Computational results of the approximate approach (continuation).
| | |||
---|---|---|---|---|
Minimum | Average | Standard Deviation | Maximum | |
| 91.94 | 92.49 | 2.74E+03 | 93.03 |
| ||||
| 91.94 | 92.49 | 2.70E+03 | 93.03 |
| ||||
| 91.94 | 92.50 | 2.68E+03 | 93.03 |
| ||||
| 91.94 | 92.50 | 2.70E+03 | 93.03 |
| ||||
| 91.94 | 92.50 | 2.70E+03 | 93.03 |
| ||||
| 91.94 | 92.51 | 2.65E+03 | 93.03 |
| ||||
| 91.94 | 92.51 | 2.60E+03 | 93.03 |
| ||||
| 91.94 | 92.51 | 2.61E+03 | 93.03 |
| ||||
| 91.94 | 92.52 | 2.60E+03 | 93.03 |
| ||||
| 91.94 | 92.52 | 2.59E+03 | 93.03 |
| ||||
| 91.94 | 92.52 | 2.59E+03 | 93.03 |
| ||||
| 91.94 | 92.53 | 2.59E+03 | 93.03 |
| ||||
| 91.94 | 92.53 | 2.59E+03 | 93.03 |
| ||||
| 91.94 | 92.53 | 2.61E+03 | 93.03 |
| ||||
| 91.94 | 92.53 | 2.62E+03 | 93.03 |
| ||||
| 91.94 | 92.53 | 2.62E+03 | 93.03 |
| ||||
| 91.98 | 92.53 | 2.59E+03 | 93.03 |
| ||||
| 91.98 | 92.53 | 2.61E+03 | 93.03 |
| ||||
| 91.98 | 92.54 | 2.58E+03 | 93.03 |
| ||||
| 91.98 | 92.54 | 2.58E+03 | 93.03 |
Computational results of the approximate approach (final).
| | |||
---|---|---|---|---|
Minimum | Average | Standard Deviation | Maximum | |
| 91.98 | 92.54 | 2.59E+03 | 93.03 |
| ||||
| 91.98 | 92.54 | 2.59E+03 | 93.03 |
| ||||
| 91.98 | 92.54 | 2.59E+03 | 93.03 |
| ||||
| 91.98 | 92.54 | 2.58E+03 | 93.03 |
| ||||
| 91.98 | 92.54 | 2.55E+03 | 93.03 |
| ||||
| 91.98 | 92.55 | 2.57E+03 | 93.03 |
| ||||
| 91.98 | 92.55 | 2.55E+03 | 93.03 |
| ||||
| 91.98 | 92.55 | 2.55E+03 | 93.03 |
| ||||
| 91.98 | 92.55 | 2.55E+03 | 93.03 |
| ||||
| 92.03 | 92.55 | 2.51E+03 | 93.03 |
| ||||
| 92.08 | 92.55 | 2.48E+03 | 93.03 |
| ||||
| 92.08 | 92.56 | 2.49E+03 | 93.03 |
| ||||
| 92.08 | 92.56 | 2.49E+03 | 93.03 |
| ||||
| 92.08 | 92.56 | 2.49E+03 | 93.03 |
| ||||
| 92.08 | 92.56 | 2.49E+03 | 93.03 |
| ||||
| 92.13 | 92.56 | 2.43E+03 | 93.03 |
| ||||
| 92.13 | 92.56 | 2.43E+03 | 93.03 |
| ||||
| 92.13 | 92.56 | 2.41E+03 | 93.03 |
| ||||
| 92.13 | 92.57 | 2.41E+03 | 93.03 |
| ||||
| 92.13 | 92.57 | 2.41E+03 | 93.03 |
In Tables
Dataset of experimental results. Twenty-five first iterations of the ten first runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 88.24 | 88.57 | 87.95 | 89.04 | 86.48 | 88.05 | 86.15 | 87.33 | 88.85 | 86.95 |
| ||||||||||
| 88.76 | 88.57 | 88.14 | 89.04 | 86.48 | 88.05 | 87.43 | 87.33 | 89.61 | 88.24 |
| ||||||||||
| 89.28 | 89.52 | 88.76 | 89.04 | 87.43 | 88.43 | 89.18 | 88.66 | 90.28 | 88.47 |
| ||||||||||
| 91.18 | 89.52 | 89.23 | 89.04 | 88.43 | 88.61 | 89.47 | 88.9 | 90.28 | 89.23 |
| ||||||||||
| 91.18 | 91.08 | 90.42 | 89.47 | 88.61 | 89.33 | 90.04 | 88.9 | 92.13 | 89.8 |
| ||||||||||
| 91.46 | 91.08 | 91.41 | 90.32 | 88.61 | 89.37 | 91.32 | 89.14 | 92.13 | 90.23 |
| ||||||||||
| 91.46 | 91.37 | 91.46 | 90.32 | 89.04 | 89.37 | 91.46 | 89.47 | 92.27 | 90.37 |
| ||||||||||
| 91.75 | 91.46 | 92.46 | 90.61 | 89.47 | 89.37 | 91.98 | 89.47 | 92.27 | 90.56 |
| ||||||||||
| 91.84 | 91.46 | 92.46 | 90.94 | 90.09 | 89.61 | 91.98 | 90.32 | 92.31 | 91.51 |
| ||||||||||
| 91.89 | 91.94 | 92.5 | 90.94 | 90.56 | 90.42 | 92.13 | 90.7 | 92.31 | 91.51 |
| ||||||||||
| 91.94 | 91.94 | 92.5 | 90.94 | 91.03 | 90.42 | 92.13 | 90.75 | 92.31 | 91.51 |
| ||||||||||
| 91.98 | 91.98 | 92.5 | 90.94 | 91.46 | 90.75 | 92.22 | 91.37 | 92.31 | 91.51 |
| ||||||||||
| 91.98 | 92.13 | 92.55 | 90.94 | 91.46 | 90.8 | 92.22 | 91.6 | 92.31 | 91.6 |
| ||||||||||
| 91.98 | 92.31 | 92.55 | 90.94 | 91.65 | 91.13 | 92.22 | 91.6 | 92.31 | 91.6 |
| ||||||||||
| 91.98 | 92.31 | 92.55 | 90.99 | 91.7 | 91.13 | 92.22 | 91.75 | 92.31 | 91.6 |
| ||||||||||
| 92.03 | 92.36 | 92.6 | 91.03 | 91.75 | 91.13 | 92.22 | 91.75 | 92.36 | 91.6 |
| ||||||||||
| 92.03 | 92.46 | 92.6 | 91.08 | 91.98 | 91.13 | 92.22 | 91.75 | 92.36 | 91.6 |
| ||||||||||
| 92.13 | 92.46 | 92.6 | 91.08 | 91.98 | 91.13 | 92.22 | 91.75 | 92.41 | 91.7 |
| ||||||||||
| 92.13 | 92.46 | 92.6 | 91.08 | 91.98 | 91.13 | 92.22 | 91.75 | 92.5 | 91.7 |
| ||||||||||
| 92.17 | 92.46 | 92.6 | 91.22 | 91.98 | 91.13 | 92.22 | 91.79 | 92.5 | 91.89 |
| ||||||||||
| 92.17 | 92.46 | 92.6 | 91.37 | 92.03 | 91.13 | 92.22 | 91.98 | 92.5 | 91.89 |
| ||||||||||
| 92.17 | 92.46 | 92.6 | 91.37 | 92.03 | 91.13 | 92.22 | 91.98 | 92.5 | 91.89 |
| ||||||||||
| 92.17 | 92.46 | 92.6 | 91.37 | 92.13 | 91.13 | 92.31 | 92.13 | 92.5 | 91.89 |
| ||||||||||
| 92.17 | 92.5 | 92.6 | 91.51 | 92.13 | 91.13 | 92.31 | 92.13 | 92.5 | 91.89 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.51 | 92.13 | 91.13 | 92.36 | 92.13 | 92.5 | 91.89 |
Dataset of experimental results. Twenty-five second iterations of the ten first runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 92.27 | 92.5 | 92.6 | 91.51 | 92.31 | 91.13 | 92.41 | 92.13 | 92.5 | 91.89 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.51 | 92.31 | 91.13 | 92.46 | 92.13 | 92.5 | 91.89 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.51 | 92.31 | 91.18 | 92.55 | 92.13 | 92.5 | 91.89 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.46 | 91.18 | 92.55 | 92.13 | 92.5 | 91.89 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.46 | 91.18 | 92.65 | 92.13 | 92.5 | 91.94 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.46 | 91.27 | 92.65 | 92.17 | 92.5 | 91.94 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.5 | 91.27 | 92.65 | 92.22 | 92.5 | 91.98 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.5 | 91.27 | 92.65 | 92.27 | 92.5 | 91.98 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.65 | 91.27 | 92.69 | 92.41 | 92.5 | 92.03 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.79 | 91.32 | 92.69 | 92.41 | 92.5 | 92.13 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.88 | 91.32 | 92.69 | 92.41 | 92.5 | 92.17 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.88 | 91.6 | 92.69 | 92.41 | 92.5 | 92.17 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.88 | 91.6 | 92.69 | 92.41 | 92.5 | 92.17 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.98 | 91.6 | 92.69 | 92.41 | 92.5 | 92.17 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.98 | 91.6 | 92.69 | 92.41 | 92.5 | 92.27 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.98 | 91.6 | 92.69 | 92.41 | 92.5 | 92.36 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.98 | 91.65 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.98 | 91.65 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.56 | 92.98 | 91.65 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.7 | 92.98 | 91.65 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.89 | 92.98 | 91.84 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.89 | 92.98 | 91.84 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.89 | 92.98 | 91.84 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.27 | 92.5 | 92.6 | 91.89 | 92.98 | 91.84 | 92.69 | 92.41 | 92.5 | 92.5 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 92.98 | 91.84 | 92.69 | 92.41 | 92.5 | 92.55 |
Dataset of experimental results. Twenty-five third iterations of the ten first runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 91.84 | 92.69 | 92.41 | 92.5 | 92.6 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.17 | 92.69 | 92.41 | 92.5 | 92.65 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.17 | 92.69 | 92.41 | 92.5 | 92.65 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.17 | 92.69 | 92.41 | 92.5 | 92.65 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.17 | 92.69 | 92.41 | 92.5 | 92.65 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.22 | 92.69 | 92.41 | 92.5 | 92.65 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.22 | 92.69 | 92.41 | 92.5 | 92.65 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.22 | 92.69 | 92.41 | 92.5 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.22 | 92.69 | 92.41 | 92.5 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.27 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 91.94 | 93.03 | 92.27 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.03 | 93.03 | 92.27 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.03 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.03 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.03 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.08 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.31 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
Dataset of experimental results. Twenty-five fourth iterations of the ten first runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.36 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.41 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.41 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.41 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.41 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.31 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 | 92.69 |
| ||||||||||
| 92.41 | 92.5 | 92.6 | 92.13 | 93.03 | 92.46 | 92.69 | 92.46 | 92.55 |
Dataset of experimental results. Twenty-five first iterations of the ten second runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 86.86 | 87 | 85.91 | 87.86 | 88.52 | 88.28 | 88.43 | 85.39 | 86.39 | 86.2 |
| ||||||||||
| 87.57 | 88.28 | 88.24 | 87.86 | 88.52 | 88.28 | 88.52 | 86.48 | 87.9 | 86.2 |
| ||||||||||
| 88.57 | 88.95 | 88.43 | 89.71 | 88.52 | 89.14 | 88.71 | 88.9 | 88.09 | 87.95 |
| ||||||||||
| 88.85 | 88.95 | 88.8 | 89.75 | 88.66 | 89.14 | 88.8 | 88.9 | 88.57 | 87.95 |
| ||||||||||
| 89.14 | 89.71 | 88.8 | 90.09 | 89.28 | 89.14 | 88.8 | 88.9 | 89.42 | 87.95 |
| ||||||||||
| 89.28 | 90.51 | 88.8 | 90.09 | 90.51 | 89.94 | 88.8 | 89.33 | 90.37 | 88.19 |
| ||||||||||
| 89.52 | 90.51 | 89.42 | 91.94 | 91.37 | 90.04 | 89.28 | 89.33 | 90.51 | 88.47 |
| ||||||||||
| 89.71 | 90.8 | 89.42 | 91.94 | 91.37 | 90.18 | 89.37 | 89.33 | 91.27 | 90.04 |
| ||||||||||
| 89.94 | 90.84 | 89.42 | 92.13 | 91.37 | 90.94 | 89.9 | 89.33 | 91.27 | 90.56 |
| ||||||||||
| 90.37 | 91.08 | 89.52 | 92.46 | 91.79 | 91.84 | 89.9 | 89.42 | 91.27 | 90.89 |
| ||||||||||
| 90.56 | 91.18 | 89.52 | 92.46 | 91.79 | 91.84 | 90.04 | 89.52 | 91.75 | 90.89 |
| ||||||||||
| 91.27 | 91.51 | 89.56 | 92.55 | 91.79 | 91.84 | 91.22 | 89.52 | 91.84 | 90.89 |
| ||||||||||
| 91.27 | 91.56 | 89.56 | 92.55 | 91.79 | 91.89 | 91.22 | 89.66 | 91.94 | 90.89 |
| ||||||||||
| 91.27 | 91.7 | 90.56 | 92.55 | 91.84 | 91.89 | 91.7 | 90.09 | 92.13 | 90.89 |
| ||||||||||
| 91.37 | 91.75 | 91.13 | 92.55 | 91.84 | 91.89 | 91.7 | 90.09 | 92.13 | 90.99 |
| ||||||||||
| 91.51 | 91.75 | 91.13 | 92.55 | 91.84 | 91.89 | 91.7 | 90.42 | 92.13 | 90.99 |
| ||||||||||
| 91.6 | 91.79 | 91.18 | 92.55 | 91.84 | 91.89 | 91.79 | 90.42 | 92.13 | 90.99 |
| ||||||||||
| 91.6 | 91.79 | 91.18 | 92.55 | 91.89 | 91.89 | 91.79 | 90.42 | 92.13 | 90.99 |
| ||||||||||
| 91.7 | 91.79 | 91.37 | 92.55 | 92.08 | 91.94 | 91.79 | 90.65 | 92.13 | 90.99 |
| ||||||||||
| 91.7 | 91.79 | 91.37 | 92.6 | 92.08 | 91.94 | 91.79 | 90.8 | 92.17 | 90.99 |
| ||||||||||
| 91.75 | 91.79 | 91.6 | 92.6 | 92.08 | 91.94 | 91.89 | 91.03 | 92.17 | 90.99 |
| ||||||||||
| 91.79 | 91.84 | 91.6 | 92.6 | 92.08 | 91.94 | 91.94 | 91.56 | 92.17 | 91.03 |
| ||||||||||
| 91.79 | 91.84 | 91.65 | 92.6 | 92.22 | 91.94 | 91.94 | 91.56 | 92.17 | 91.27 |
| ||||||||||
| 91.79 | 91.94 | 91.65 | 92.6 | 92.22 | 91.94 | 91.94 | 91.6 | 92.17 | 91.41 |
| ||||||||||
| 91.79 | 91.94 | 91.65 | 92.6 | 92.22 | 91.94 | 91.94 | 91.6 | 92.17 | 91.41 |
Dataset of experimental results. Twenty-five second iterations of the ten second runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 91.79 | 91.94 | 91.65 | 92.6 | 92.22 | 91.94 | 91.94 | 91.98 | 92.17 | 91.41 |
| ||||||||||
| 91.79 | 91.94 | 91.65 | 92.6 | 92.22 | 91.94 | 91.94 | 91.98 | 92.17 | 91.41 |
| ||||||||||
| 91.79 | 91.94 | 91.75 | 92.6 | 92.22 | 91.98 | 91.94 | 92.03 | 92.17 | 91.41 |
| ||||||||||
| 91.84 | 91.94 | 91.79 | 92.6 | 92.22 | 91.98 | 91.94 | 92.03 | 92.17 | 91.41 |
| ||||||||||
| 91.84 | 91.94 | 91.89 | 92.6 | 92.22 | 92.03 | 91.94 | 92.08 | 92.17 | 91.65 |
| ||||||||||
| 91.84 | 91.94 | 91.89 | 92.6 | 92.22 | 92.03 | 91.94 | 92.17 | 92.17 | 91.65 |
| ||||||||||
| 91.84 | 91.94 | 91.89 | 92.6 | 92.22 | 92.03 | 91.94 | 92.27 | 92.17 | 91.65 |
| ||||||||||
| 91.84 | 91.94 | 91.94 | 92.6 | 92.22 | 92.03 | 91.94 | 92.27 | 92.17 | 91.65 |
| ||||||||||
| 91.84 | 91.94 | 91.94 | 92.6 | 92.22 | 92.08 | 91.94 | 92.27 | 92.17 | 91.65 |
| ||||||||||
| 91.84 | 91.94 | 91.94 | 92.6 | 92.22 | 92.08 | 91.94 | 92.27 | 92.17 | 91.7 |
| ||||||||||
| 91.89 | 91.94 | 91.94 | 92.6 | 92.22 | 92.13 | 91.94 | 92.27 | 92.17 | 91.7 |
| ||||||||||
| 91.89 | 91.94 | 91.94 | 92.6 | 92.27 | 92.17 | 91.98 | 92.31 | 92.17 | 91.75 |
| ||||||||||
| 91.89 | 91.94 | 91.94 | 92.6 | 92.27 | 92.22 | 92.08 | 92.36 | 92.17 | 91.75 |
| ||||||||||
| 91.89 | 91.94 | 91.94 | 92.69 | 92.27 | 92.22 | 92.17 | 92.36 | 92.17 | 91.94 |
| ||||||||||
| 91.89 | 91.94 | 91.94 | 92.69 | 92.27 | 92.27 | 92.17 | 92.36 | 92.17 | 91.94 |
| ||||||||||
| 91.89 | 91.94 | 91.94 | 92.74 | 92.27 | 92.36 | 92.17 | 92.36 | 92.17 | 92.13 |
| ||||||||||
| 92.03 | 91.94 | 91.98 | 92.74 | 92.27 | 92.36 | 92.31 | 92.36 | 92.17 | 92.13 |
| ||||||||||
| 92.03 | 91.94 | 91.98 | 92.74 | 92.27 | 92.36 | 92.41 | 92.36 | 92.17 | 92.17 |
| ||||||||||
| 92.08 | 91.94 | 91.98 | 92.74 | 92.27 | 92.36 | 92.5 | 92.36 | 92.17 | 92.17 |
| ||||||||||
| 92.08 | 91.94 | 92.03 | 92.74 | 92.27 | 92.36 | 92.5 | 92.36 | 92.17 | 92.27 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.74 | 92.27 | 92.36 | 92.5 | 92.36 | 92.22 | 92.46 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.36 | 92.5 | 92.36 | 92.22 | 92.55 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.41 | 92.5 | 92.36 | 92.31 | 92.55 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.41 | 92.5 | 92.36 | 92.31 | 92.6 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.46 | 92.5 | 92.36 | 92.36 | 92.79 |
Dataset of experimental results. Twenty-five third iterations of the ten second runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.46 | 92.5 | 92.36 | 92.36 | 92.79 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.5 | 92.5 | 92.36 | 92.36 | 92.79 |
| ||||||||||
| 92.27 | 91.94 | 92.03 | 92.84 | 92.27 | 92.5 | 92.5 | 92.36 | 92.46 | 92.84 |
| ||||||||||
| 92.27 | 91.94 | 92.08 | 92.84 | 92.27 | 92.5 | 92.5 | 92.36 | 92.5 | 92.84 |
| ||||||||||
| 92.31 | 91.94 | 92.08 | 92.84 | 92.27 | 92.55 | 92.5 | 92.36 | 92.69 | 92.84 |
| ||||||||||
| 92.31 | 91.94 | 92.08 | 92.88 | 92.27 | 92.55 | 92.5 | 92.36 | 92.74 | 92.93 |
| ||||||||||
| 92.31 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.74 | 92.93 |
| ||||||||||
| 92.36 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.74 | 92.93 |
| ||||||||||
| 92.36 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.79 | 92.93 |
| ||||||||||
| 92.36 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.79 | 92.93 |
| ||||||||||
| 92.36 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.79 | 92.93 |
| ||||||||||
| 92.36 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.84 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.84 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.41 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.46 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.46 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.46 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.5 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.5 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
Dataset of experimental results. Twenty-five fourth iterations of the ten second runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 92.41 | 91.94 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.13 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.41 | 91.98 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.88 | 92.93 |
| ||||||||||
| 92.46 | 91.98 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 91.98 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 91.98 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 91.98 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.03 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.08 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.55 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.08 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.08 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.08 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.08 | 92.17 | 92.93 | 92.27 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.17 | 92.17 | 92.93 | 92.31 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.17 | 92.17 | 92.93 | 92.31 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.17 | 92.17 | 92.93 | 92.36 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.17 | 92.17 | 92.93 | 92.41 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
| ||||||||||
| 92.46 | 92.17 | 92.17 | 92.93 | 92.41 | 92.55 | 92.5 | 92.65 | 92.93 | 92.93 |
Dataset of experimental results. Twenty-five first iterations of the ten third runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 85.34 | 87.81 | 89.33 | 88.14 | 87.95 | 87.76 | 88.85 | 88.66 | 86.2 | 86.39 |
| ||||||||||
| 86.67 | 88.95 | 89.33 | 88.14 | 87.95 | 87.76 | 89.61 | 88.66 | 87.81 | 88.33 |
| ||||||||||
| 88.14 | 88.95 | 89.71 | 88.61 | 87.95 | 89.66 | 90.28 | 88.66 | 88.8 | 88.61 |
| ||||||||||
| 88.47 | 88.95 | 90.51 | 90.04 | 88.33 | 89.66 | 90.28 | 88.8 | 88.85 | 88.66 |
| ||||||||||
| 88.61 | 88.95 | 91.94 | 90.04 | 88.52 | 89.66 | 92.13 | 88.85 | 89.28 | 88.8 |
| ||||||||||
| 88.66 | 88.95 | 91.98 | 90.04 | 88.61 | 90.32 | 92.13 | 88.9 | 89.37 | 88.99 |
| ||||||||||
| 88.66 | 89.04 | 92.03 | 90.56 | 89.47 | 90.32 | 92.27 | 89.47 | 90.23 | 88.99 |
| ||||||||||
| 88.66 | 89.33 | 92.17 | 91.18 | 89.75 | 90.56 | 92.27 | 89.47 | 90.84 | 88.99 |
| ||||||||||
| 89.47 | 89.66 | 92.17 | 91.89 | 90.04 | 91.46 | 92.31 | 89.75 | 91.13 | 89.09 |
| ||||||||||
| 89.47 | 90.04 | 92.17 | 91.89 | 91.03 | 91.46 | 92.31 | 90.32 | 91.13 | 89.23 |
| ||||||||||
| 89.61 | 91.6 | 92.17 | 92.13 | 91.08 | 91.46 | 92.31 | 90.42 | 91.7 | 89.56 |
| ||||||||||
| 90.89 | 91.6 | 92.17 | 92.17 | 91.13 | 91.46 | 92.31 | 90.89 | 91.84 | 90.09 |
| ||||||||||
| 90.89 | 91.84 | 92.17 | 92.31 | 91.32 | 91.6 | 92.31 | 91.79 | 91.84 | 90.09 |
| ||||||||||
| 90.89 | 92.08 | 92.22 | 92.31 | 91.32 | 91.6 | 92.31 | 91.98 | 91.84 | 90.09 |
| ||||||||||
| 90.99 | 92.08 | 92.22 | 92.36 | 91.37 | 91.79 | 92.31 | 91.98 | 91.89 | 90.09 |
| ||||||||||
| 90.99 | 92.08 | 92.22 | 92.36 | 91.46 | 91.98 | 92.36 | 92.17 | 91.94 | 90.32 |
| ||||||||||
| 91.13 | 92.27 | 92.22 | 92.36 | 91.46 | 91.98 | 92.36 | 92.17 | 91.94 | 90.8 |
| ||||||||||
| 91.13 | 92.27 | 92.22 | 92.36 | 91.46 | 92.27 | 92.41 | 92.17 | 91.94 | 91.03 |
| ||||||||||
| 91.13 | 92.46 | 92.22 | 92.36 | 91.46 | 92.27 | 92.5 | 92.22 | 91.94 | 91.08 |
| ||||||||||
| 91.13 | 92.46 | 92.22 | 92.36 | 91.56 | 92.27 | 92.5 | 92.27 | 91.94 | 91.08 |
| ||||||||||
| 91.13 | 92.5 | 92.22 | 92.36 | 92.03 | 92.27 | 92.5 | 92.27 | 91.94 | 91.08 |
| ||||||||||
| 91.13 | 92.55 | 92.22 | 92.36 | 92.03 | 92.27 | 92.5 | 92.31 | 91.94 | 91.13 |
| ||||||||||
| 91.13 | 92.6 | 92.22 | 92.41 | 92.08 | 92.27 | 92.5 | 92.31 | 91.94 | 91.27 |
| ||||||||||
| 91.13 | 92.79 | 92.27 | 92.46 | 92.08 | 92.27 | 92.5 | 92.31 | 91.94 | 91.41 |
| ||||||||||
| 91.13 | 92.79 | 92.27 | 92.46 | 92.17 | 92.27 | 92.5 | 92.36 | 91.94 | 91.41 |
Dataset of experimental results. Twenty-five second iterations of the ten third runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 91.13 | 92.79 | 92.31 | 92.5 | 92.17 | 92.27 | 92.5 | 92.36 | 91.94 | 91.51 |
| ||||||||||
| 91.13 | 92.79 | 92.31 | 92.5 | 92.17 | 92.27 | 92.5 | 92.46 | 91.94 | 91.65 |
| ||||||||||
| 91.13 | 92.79 | 92.31 | 92.5 | 92.17 | 92.27 | 92.5 | 92.46 | 91.94 | 91.7 |
| ||||||||||
| 91.13 | 92.84 | 92.31 | 92.5 | 92.17 | 92.27 | 92.5 | 92.5 | 91.94 | 91.79 |
| ||||||||||
| 91.13 | 92.84 | 92.31 | 92.5 | 92.31 | 92.31 | 92.5 | 92.5 | 91.98 | 91.79 |
| ||||||||||
| 91.13 | 92.93 | 92.36 | 92.5 | 92.36 | 92.36 | 92.5 | 92.5 | 91.98 | 91.84 |
| ||||||||||
| 91.18 | 92.93 | 92.36 | 92.5 | 92.36 | 92.36 | 92.5 | 92.5 | 91.98 | 91.84 |
| ||||||||||
| 91.22 | 92.93 | 92.36 | 92.5 | 92.41 | 92.36 | 92.5 | 92.5 | 91.98 | 91.84 |
| ||||||||||
| 91.27 | 92.93 | 92.36 | 92.5 | 92.46 | 92.36 | 92.5 | 92.5 | 91.98 | 91.89 |
| ||||||||||
| 91.41 | 92.93 | 92.36 | 92.5 | 92.46 | 92.36 | 92.5 | 92.5 | 91.98 | 91.98 |
| ||||||||||
| 91.56 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.5 | 91.98 | 92.08 |
| ||||||||||
| 91.56 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.5 | 91.98 | 92.13 |
| ||||||||||
| 91.75 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.5 | 91.98 | 92.13 |
| ||||||||||
| 91.75 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.55 | 92.08 | 92.13 |
| ||||||||||
| 91.75 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.55 | 92.08 | 92.17 |
| ||||||||||
| 91.89 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.55 | 92.13 | 92.17 |
| ||||||||||
| 91.89 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.55 | 92.13 | 92.17 |
| ||||||||||
| 91.89 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.55 | 92.17 | 92.17 |
| ||||||||||
| 91.89 | 92.93 | 92.36 | 92.5 | 92.5 | 92.36 | 92.5 | 92.55 | 92.27 | 92.17 |
| ||||||||||
| 91.94 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.36 | 92.17 |
| ||||||||||
| 91.94 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.41 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.41 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
Dataset of experimental results. Twenty-five third iterations of the ten third runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 91.98 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.17 |
| ||||||||||
| 92.03 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.22 |
| ||||||||||
| 92.03 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.5 | 92.55 | 92.5 | 92.27 |
| ||||||||||
| 92.08 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.41 |
| ||||||||||
| 92.08 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.41 |
| ||||||||||
| 92.08 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.41 |
| ||||||||||
| 92.08 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.41 |
| ||||||||||
| 92.08 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.41 |
| ||||||||||
| 92.08 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.41 |
| ||||||||||
| 92.13 | 92.93 | 92.36 | 92.5 | 92.5 | 92.41 | 92.55 | 92.55 | 92.5 | 92.6 |
| ||||||||||
| 92.17 | 92.93 | 92.36 | 92.5 | 92.55 | 92.41 | 92.55 | 92.55 | 92.5 | 92.6 |
| ||||||||||
| 92.17 | 92.93 | 92.36 | 92.5 | 92.55 | 92.41 | 92.55 | 92.55 | 92.5 | 92.65 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.6 | 92.41 | 92.55 | 92.55 | 92.5 | 92.69 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.6 | 92.41 | 92.55 | 92.55 | 92.5 | 92.69 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.6 | 92.41 | 92.55 | 92.55 | 92.5 | 92.69 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.65 | 92.41 | 92.55 | 92.55 | 92.5 | 92.69 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.65 | 92.41 | 92.55 | 92.55 | 92.5 | 92.69 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.65 | 92.41 | 92.55 | 92.55 | 92.5 | 92.74 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.65 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
Dataset of experimental results. Twenty-five fourth iterations of the ten third runs.
| | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
| | | | | | | | | | |
| 92.22 | 92.93 | 92.36 | 92.5 | 92.65 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.65 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.22 | 92.93 | 92.36 | 92.5 | 92.74 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.74 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.74 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.79 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.84 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.84 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.5 | 92.84 | 92.41 | 92.55 | 92.55 | 92.5 | 92.79 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.55 | 92.84 | 92.41 | 92.55 | 92.55 | 92.5 | 92.84 |
| ||||||||||
| 92.27 | 92.93 | 92.36 | 92.55 | 92.84 | 92.41 | 92.55 | 92.55 | 92.5 | 92.84 |
The software developed and the data generated to support the findings of this study have been deposited in the Figshare repository (10.6084/m9.figshare.5588896, 10.6084/m9.figshare.5588911, and 10.6084/m9.figshare.5590000.v2).
The authors declare that there are no conflicts of interest regarding the publication of this paper.
Roberto Munoz and Rodrigo Olivares are supported by Postgraduate Grant of Pontificia Universidad Católica de Valparaíso (INF-PUCV 2015). Carla Taramasco and Rodrigo Olivares are supported by CONICYT/FONDEF/IDeA/ID16I10449, CONICYT/STIC-AMSUD/17STIC-03, and CONICYT/MEC/MEC80170097 and CENS (National Center for Health Information Systems). Rodolfo Villarroel is funded by the VRIEA-PUCV 2017 039.440/2017 Grant. Ricardo Soto is supported by Grant CONICYT/FONDECYT/REGULAR/1160455. María Francisca Alonso-Sánchez is supported by CONICYT, FONDECYT INICIACION 11160212. Roberto Munoz and Carla Taramasco also acknowledge the Center for Research and Development in Health Engineering of the Universidad de Valparaíso. Finally, the authors would like to thank Travis Jones for his valuable contributions to the elaboration of this paper.