A MODIFIED MAXIMUM RELEVANCE MINIMUM REDUNDANCY FEATURE SELECTION METHOD BASED ON TABU SEARCH FOR PARKINSON’S DISEASE MINING

International Journal of Artificial Intelligence and Applications (IJAIA), Vol.11, No.2, March 2020
DOI: 10.5121/ijaia.2020.11201 1
A MODIFIED MAXIMUM RELEVANCE MINIMUM
REDUNDANCY FEATURE SELECTION
METHOD BASED ON TABU SEARCH FOR
PARKINSON’S DISEASE MINING
Waheeda Almayyan
Computer Information Department, Collage of Business Studies, PAAET, Kuwait
ABSTRACT
Parkinson’s disease is a complex chronic neurodegenerative disorder of the central nervous system. One of
the common symptoms for the Parkinson’s disease subjects, is vocal performance degradation. Patients
usually advised to follow personalized rehabilitative treatment sessions with speech experts. Recent
research trends aim to investigate the potential of using sustained vowel phonations for replicating the
speech experts’ assessments of Parkinson’s disease subjects’ voices. With the purpose of improving the
accuracy and efficiency of Parkinson’s disease treatment, this article proposes a two-stage diagnosis
model to evaluate an LSVT dataset. Firstly, we propose a modified minimum Redundancy-Maximum
Relevance (mRMR) feature selection approach, based on Cuckoo Search and Tabu Search to reduce the
features numbers. Secondly, we apply simple random sampling technique to dataset to increase the
samples of the minority class. Promisingly, the developed approach obtained a classification Accuracy rate
of 95% with 24 features by 10-fold CV method.
KEYWORDS
Parkinson’s disease; Medical data mining; maximum Relevance Minimum Redundancy, Tabu Search&
Simple Random Sampling.
1. INTRODUCTION
Parkinson’s disease (PD) is a complex progressive, degenerative universal disorder of the central
nervous system. PD is the second most common neurodegenerative disorder in elderly.
According to the World Health Organization (WHO), it was estimated that there are about 7-10
million PD patients in the world [1]. Two percent of the population above the age of 60 is
affected by the PD. It ultimately leads to several motor and non-motor manifestations. The cause
of PD is still unknown and early signs may be mild and unnoticeable, the diagnoses of PD disease
mainly relies on clinical criteria [2]. Research has shown that approximately 90 percent of PD
patients face some form of vocal deficiency [3].
Classification systems play a major role in machine learning and knowledge-based systems. In
the last 30 years Computer-Aided Diagnosis (CAD) has become one of the vital research topics
in medical diagnostic tasks. Medical practitioners depend upon the experience beside the existing
information, whereas, CAD usually apply intelligent machine learning techniques to help
physicians in making decisions. A number of published articles suggested several strategies to
process the physician’s interpretation and decisions [4]. During classification process, accuracy

2
of clinically diagnosed cases is particularly important issue to be considered. But due to the
voluminous and heterogeneous nature of medical data: classification techniques are required to
make accurate decisions. The size of most medical datasets is normally large-scaled; accordingly,
this matter effects the complexity of the data mining process [5].Therefore, dimension reduction
techniques play a major role in excluding irrelevant features from medical datasets [6]. Therefore,
instead of using all existing features, dimension reduction procedure aims to reduce
computational complexity with the possible advantages of enhancing the overall classification
performance and eventually reduce the computational cost.
This article proposes a two-stage feature selection algorithm to assess the PD habitation speech
treatment based on Tabu and minimum redundancy - maximum relevance (mRMR) search
techniques. Firstly, we select the first set of features by applying Tabu Search to construct a first
reductive feature set. Secondly, we apply a modified mRMR feature selection approach, based on
Cuckoo Search to further reduce the dimension and to discover the important relationship
between features. Finally, we apply the random sampling technique to handle the imbalanced
distribution of the medical data. This paper attempts to extend the work of [7] to deal with the
limited volume of data and improve the accuracy of the prediction event.
The reminder of this article is organized in the following manner; introduction and related
research are briefly described in sections 1 and 2. Section 3 explains the theoretical approach of
feature selection methods and the proposed technique. The dataset, evaluation procedure and the
experimental results are presented in sections 4 and 5. Finally, conclusion is summarized in
Section 6.
2. RELATED WORK
Several studies have been carried out by numerous researchers on the early diagnosis of PD based
on machine learning methods [8-20]. Little and his co-researchers conducted an important study
to detect PD by introducing a new measure of dysphonia [8]. This investigation has presented a
new dysphonia measure, known as the pitch period entropy. They employed Support Vector
Machine (SVM) classifier with Gaussian radial basis kernel functions to predict PD and
performed feature selection to select the optimal subset of features. Their best overall accuracy
rate was 91.4%. In 2010, Das compare Neural Networks (ANN), DMneural, Regression and
Decision Tree to predict PD. The results have shown that the ANN classifier approach
outperformed the other two models with an overall accuracy of 92.9% [9]. Authors in [10]
introduced a nonlinear model based on Dirichlet mixtures. The experimental results showed that
the suggested model outperformed the multinomial logit models, decision trees and SVM with an
overall accuracy of 87.7%.
Sakar and Kursun (2010) selected a minimal subset of dysphonia features according to the
Maximum Relevance Minimum Redundancy (mRMR) criterion. An accuracy of 92.75% with the
SVM classifier was obtained [11]. Psorakis et al. (2010) introduced sample selection strategies,
novel convergence measures and model improvements for multiclass multi-kernel relevance
vector machines (mRVMs), and finally, the improved mRVMs achieved the classification
accuracy rate of 89.47% [12]. Researchers in [13] built a classification model based on genetic
programming and the expectation maximization algorithm to detect PD. According to the
experimental results the best classification accuracy obtained during the experiments was 93.1%.
Luukka (2011) suggested a feature selection model based on fuzzy entropy measures together
with the similarity classifier to predict PD and an additional three medical datasets. The mean
classification accuracy of predicting PD was 85.03% with only two features from original 22
features [14].

3
Authors in [15] achieved an accuracy of 93.47% with a model based on non-linear fuzzy
transformation method in combination with the principal component analysis together with SVM
classifier. Ozcift and Gulten (2011) started with reducing feature set with the correlation-based
feature selection (CFS) algorithm. Then, they applied the classification of the rotation forest
ensemble classifiers that composed of 30 machine-learning algorithms to achieve an average
accuracy of 87.13% [16]. For the prediction of PD, AStröm and Koker (2011) proposed 9 parallel
feed-forward NN structure. The output of each neural network was evaluated using a rule-based
system to make final decision. Moreover, in the proposed approach unlearned data of each neural
network during the training process is collected and used in the training set of the next neural
network. The highest reported classification accuracy was 91.20% [17]. Researchers in [18]
applied evolutionary-based techniques in combination with the Optimum-Path Forest classifier to
reduce the number of features before detecting PD. The best classification accuracy of 84.01%
was reported using Gravitational Search Algorithm technique with eight features. Sriram et al.
(2015) deployed Sieve multigram data and Survey graph to obtain the statistical analysis on the
voice data. Then they performed clustering using the KStar and NNge classifiers [19]. Recently,
with the help of data mining and bioinformatics analysis, several potential therapeutic drugs that
may be used for prevention and treatment of Parkinson’s disease were discovered [20]. This
certainly open new perspectives for drugs development using AI techniques.
From this quick review, we can notice that most of the well-known classifiers models from
machine learning have been utilized to improve the technological assessments of PD. So,
choosing an excellent classifier is of significant importance to the tackle the PD classification
problem. Also, we have noticed that much work is needed to tackle the imbalanced nature of
medical datasets. Therefore, we will study the impact of the data pre-processing strategy that uses
data sampling after applying several attribute selection techniques, on the classification models
constructed. Aiming at improving the efficiency and effectiveness of the classification accuracy
for PD diagnosis with minimum number of features, in this article, a classification system based
on two-level of feature extraction is introduced. In this work we will focus on choosing the
appropriate features to describe the feature set based on Random Forest as a base classifier.
3. THE PROPOSED APPROACH
Experts address the rapidly rising number of deaths from PD. New study predicted that by the
year 2030 there will be around 1.2 million people living with PD in the United States [21]. A
practically collected LSVT dataset [8] is used in this article to demonstrate the proposed
procedure. More details on the dataset will be is introduced in the next section. The proposed
method for the purpose of diagnosis of PD applied in this study is illustrated in Figure 1. The
main objective of the proposed approach is to explore the performance of PD diagnosis using a
two-stage hybrid modelling procedure through integrating mRMR with Tabu Search technique.
Firstly, the proposed method adopts mRMR to construct the discriminative feature space through
Cuckoo search algorithm, then the Tabu technique is applied to help in improving the
discrimination ability of classifiers. In the last step the resulted feature space is processed by the
random resampling technique and ready to be evaluated with different types of activation
functions to perform the classification.
Eventually, five machine-learning classification methods, which are considered very robust in
solving non-linear problems, are chosen to estimate the PD possibility. These methods include
Random Forest (RF), Fuzzy Unordered Rule Induction Algorithm (FURIA), AdaBoost, Rough
sets and C4.5. In order to evaluate the prediction performance of the suggested model, we used
six performance metrics, Sensitivity, Specificity, Accuracy, Precision, 𝑓-measure, and MCC, to
test the classification performance.

4
Figure.1. The proposed methodology
3.1. mRMR feature selection method based on Cuckoo search
The mRMR method helps in extracting crucial features and this can minimize the classification
error [22]. The mRMR is a heuristic technique approach proposed by Peng et al. to measure the
relevancy and redundancy of features and determine the informative features [23]. Several
experimental results denoted that mRMR is an effective method to improve the performance of
feature selection [24,25]. The mRMR feature selection technique is meant to identify a set of
crucial features that have maximum relevancy for target classes and minimum redundancy with
other features in the dataset at the same time.
The basic concept of mRMR method is to use two mutual information MI operations: one
between classes and each feature in order to measure the relevancy, while the second mutual
information between every two features to evaluate the redundancy. 𝑆 denotes the selected
features and Rl measures the relevancy of a group of selected features 𝑆 that can be defined as
follows
𝑅𝑙 =
1
|𝑆|
∑ 𝐼( 𝐺𝑥, 𝐶)𝐺 𝑥∈𝑆 1
where 𝐼(𝐺𝑥,𝐶) denoted the value of mutual information between an individual feature 𝐺𝑥 that
belongs to 𝑆 and the target class 𝐶. When the selected features have the maximum relevance Rl
value, it is possible to have high redundancy between these features. Hence, the redundancy Rd
of a group of selected features 𝑆 is defined as
𝑅𝑑 =
1
|𝑆|2
∑ 𝐼(𝐺 𝑥, 𝐺 𝑦)𝐺 𝑥,𝐺 𝑦∈𝑆 2
Where |S| is the number of feature in S and I(𝐺𝑥 ,𝐺𝑦) is the mutual information between the 𝑥th
and 𝑦th features that measures the mutual dependency of these two features.
LSVTDataset
1st Feature Selection
Procedure
2nd Feature Selection
Procedure
Resample Data
Classification
Procedure
Performance
Measures
Sensitivity
Specificity
Precision
Accuracy
MCC
𝑓-measure
RF
FURIA
Ada Boost
C4.5
Rough Sets
Random Resample
mRMR
Tabu

5
The main purpose of applying the mRMR feature selection method is to find a subset of features
from 𝑆 with 𝑚 features, { 𝑥𝑖}, that either jointly have the largest dependency on the target class C
or have the minimal redundancy on the selected features subset 𝑆. Thus, Peng et al. [23]
recommend searching for balanced solutions through the composite objective. This criterion
combines the two criteria, which are maximal relevance criterion and minimal redundancy
criterion, as follows
max (Rl, Rd) = Rl – Rd 3
Our goal is to increase the prediction accuracy and reduce the number of selected features.
Thereafter, features are selected one by one by applying Cuckoo search to maximize the objective
function, which is a function of relevance and redundancy. Hence, we applied the mRMR method
to filter irrelevant and noisy features and eventually reduces the computational load.
3.2. Tabu Search
Tabu Search is a memory-based metaheuristic algorithm proposed by Glover in 1986 to solve
combinatorial optimization problems [26,27]. Since then, Tabu Search has been successfully
applied in other feature selection problems [28]. This search technique is a local neighborhood
search algorithm that simulates the optimal characteristics of human memory functions. Tabu
Search involves a local search combined with a tabu mechanism. It starts with an initial feasible
solution X’ among the neighborhood solutions, where  is the set of feasible solutions, and at
each iteration, the algorithm searches the neighborhood of the best solution N(X) to obtain a
new one with an improved functional value. A solution X’  N(X) can be reached from X in
two cases, X’ is not included in the tabu list; and X’ is included in the tabu list, but it satisfies the
aspiration criterion [29]. Surely, if the new solution 𝑋’ is superior to 𝑋best, the value of 𝑋best is
overridden. To avoid cycling, solutions that were previously visited are declared forbidden or
tabu for a certain number of iterations and this surely improve the performance of the local
search. Then, the neighborhood search is resumed based on the new feasible solution 𝑋’. This
whole procedure is iteratively executed until the stopping criteria is met. After the iterative
process has terminated, the current best solution so far 𝑋best is considered the final optimal
solution provided by the Tabu Search method [30].
Begin
Start with an initial feasible solution X
Initialize tabu list and aspiration level
For fixed number of iterations Do
Generate neighbour solutions
Find best 𝑋best
If move from X to X’ is not in tabu list Then
Accept move and update solution
Update tabu list and aspiration level
Increment iteration number
Else
If Cost(X’)< aspiration level
Accept move and update best solution
Update tabu list and aspiration level
Increment iteration number
End If
End If
End For
End
Pseudocode 1: Pseudocode of of Tabu Search (TS) (Sait and Youssef, 1999[31]).

6
3.3. Simple Random Sampling
Class imbalance problems occurs when one class is represented by a significantly larger number
of instances than other classes. Consequently, classification algorithms tend to ignore the
minority classes. Simple random sampling has been recommended as a method to increase the
sensitivity of the classification algorithm to the minority class by scaling the class distribution
[32,33]. Changing the class distribution can be conducted via different resampling strategies.
However, the simple random sampling technique has gained extra attention. The advantage of a
such technique is that it is external and therefore, easily transportable as well as very simple to
implement [34]. Moreover, over-sampling the minority class data avoids unnecessary information
loss [35].
Weiss and Provost conducted an empirical study where the authors used twenty datasets from
UCI repository has showed quantitatively that classifier accuracy might be increased with a
progressive sampling algorithm [32]. They used decision trees to evaluate classification
performances with the use of a sampling strategy. Another important study used simple random
sampling to scale the class distribution of biomedical datasets [33]. The authors measure the
effect of the suggested sampling strategy by the use of nearest neighbour and decision tree
classifiers. In simple random sampling, a sample is randomly selected from the population so that
the obtained sample is representative of the population. Therefore, this technique provides an
unbiased sample from the original data.
Regarding simple random sampling there are two approaches while making random selection, in
the first approach the samples are selected with replacement where the sample can be selected
more than once repeatedly with an equal selection chance. In the other approach the selection of
samples is done without replacement where the sample can be selected only once, so that each
sample in the dataset has an equal chance of being selected and once selected it cannot be chosen
again [36].
4. DATASET AND EVALUATION PROCEDURE
4.1. LSVT Dataset
Tracking the early Parkinson's disease symptom progression using the speech disorders has
shown a great sign in the advancement of Parkinson Disease detection. About 90% of people with
Parkinson’s disease present some kind of vocal deterioration. In this article, we use the dataset
originally collected by Tsanas et al. [7] to analyze the impact of LSVT (Lee Silverman Voice
Treatment) in the voice rehabilitation of 14 patients with PD. LSVT is an intensive speech
therapy program designed to improve respiratory, laryngeal, and articulatory functions during
speech in patients with Parkinson's disease [37]. Tsanas et al. measured 309 dysphonia features to
evaluate whether a sustained phonation is “acceptable” or “unacceptable” according to the
clinical criteria of six experts. Among the 126 phonations, 42 is labeled as “acceptable” while the
remaining 84 is labeled as “unacceptable”. The authors reported a classification score of 90%
considering a voting scheme. RF and SVM.
Table 1 describes the class distribution, which clearly shows that the dataset is imbalanced. Out
of 84phonations, 66.67% classified as unacceptable. A common problem with the imbalanced
data is that the minority class contributes very little to the standard algorithm accuracy.

7
Table 1.Class distribution of the LSVT dataset
Index Class name Class size Class distribution
(%)
1 acceptable 42 33.33
2 unacceptable 84 66.67
4.2. Performance Analysis
When learning of imbalanced data, the measures such as Accuracy and Error Rate usually favors
the majority class [38]. By using the 10-fold cross validation procedure, the accuracy of the
selected classifiers was calculated using sensitivity, specificity, accuracy, precision, 𝑓-measure
and MCC which are more appropriate measures for imbalanced datasets [39]. The main
formulations are defined in Equations 4-10, according to the confusion matrix, which is shown in
Table 2. In the confusion matrix of a two-class problem, TP is the number of true positives that
represent in our case the cases with Parkinson’s disease that was classified correctly. FN is the
number of false negatives that represents the cases with Parkinson’s disease that was classified
incorrectly as healthy. TN is the number of true negatives, which represents the healthy cases that
was classified as healthy. FP is the number of false positives that represents the healthy cases that
was classified as Parkinson’s disease cases.
Table 2. The confusion matrix
Hypothesis Predicted patient state
Classified case with
Parkinson’s disease
Classified healthy case
Hypothesis positive
Parkinson’s disease case
True Positive
TP
False Negative
FN
Hypothesis negative Healthy
case
False Positive
FP
True Negative
TN
The ROC (Receiver Operating Characteristic) curve is a graphical representation of the
sensitivity versus the specificity index for a classifier varying the discrimination threshold value.
The ROC curve is a standard tool for summarizing classifier performance over a range of trade-
offs between TP and FP error rates [40]. ROC usually takes values between 0.5 for random
drawing and 1.0 for perfect classifier performance.
Accuracy=
𝑇𝑃+𝑇𝑁
𝑇𝑃+𝐹𝑃+𝐹𝑁+𝑇𝑁
4
The 𝑓-measure is therefore used as it measures the harmonic mean of the classifier’s precision
and Recall as
𝑓-measure =2 ×
𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛×𝑅𝑒𝑐𝑎𝑙𝑙
𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛+𝑅𝑒𝑐𝑎𝑙𝑙
5
Accordingly, we can define Precision and Recall as
Precision=
𝑇𝑁
𝐹𝑃+𝑇𝑁
6
Recall=
𝑇𝑃
𝑇𝑃+𝐹𝑁
7

8
Considering class-imbalanced datasets, the Matthews Correlation Coefficient (MCC) is an
appropriate measure that considered balanced. It can be used even if the classes are of very
different in sizes, as it is a correlation coefficient between the observed and predicted
classification decisions. The MCC measure falls within the range of [−1,1]. The larger the MCC
coefficient indicates better classifier prediction. The MCC measure can be calculated directly
from the confusion matrix using the following formula
MCC=
𝑇𝑃.𝑇𝑁−𝐹𝑁.𝐹𝑃
√(𝑇𝑃+𝐹𝑁)(𝑇𝑃+𝐹𝑃)(𝑇𝑁+𝐹𝑁)(𝑇𝑁+𝐹𝑃)
8
Sensitivity and specificity measures can be also used to improve interpretability as follows
Sensitivity=
𝑇𝑃
𝑇𝑃+𝐹𝑁
9
Specificity=
𝑇𝑁
𝑇𝑁+𝐹𝑃
10
5. RESULTS AND DISCUSSION
One of the interesting aspects in biomedical data mining is to build computational models with
abilities to extract hidden knowledge using data mining schemes. The number of trees for RF
algorithm and decision trees for Adaboost was arbitrarily set to 100, since it has been shown that
the optimal number of trees is usually 64–128, while further increasing the number of trees does
not necessarily improve the model’s performance [41]. The MINNO (minimum total weight of
the instances in a rule) of the FURIA algorithm has been set to 2.0. The performance results are
the average after 10 runs of 10 folds-validation on each classifier to obtain its prediction
measures. The suggested model for the purpose of evaluating LSVT data applied in this study is
carried out in two major phases. In order to verify the effectiveness of the proposed model, firstly
we compare with the performance of the selected classifier on the original feature space. Table 3,
shows the classification results on the original feature space. It can be observed that the
classification Accuracy fluctuates between 75% and 84% with the whole feature set. RF has
achieved the results of 94%, 64.3%, 84%, 88.8% and 66.8% in terms of Sensitivity, Specificity,
Precision, 𝑓-measure and MCC.
Table 3. Performance measures on the original features of LSVT dataset
Classifier Sensitivi
ty
Specifici
ty
Accurac
y
Precision f-
measure
MCC
RF 0.940 0.643 0.841 0.840 0.888 0.668
FURIA 0.940 0.619 0.833 0.832 0.883 0.651
RoughSet 0.917 0.714 0.849 0.865 0.890 0.672
AdaBoost 0.833 0.595 0.754 0.805 0.819 0.445
C4.5 0.845 0.571 0.754 0.798 0.821 0.444

9
Table 4. Selected features of LSVT dataset
FS technique Number
of selected
features
Selected features labels
1. Tabu 37 Jitter->F0_TKEO_prc75
Jitter->pitch_TKEO_std
Shimmer-
>Ampl_TKEO_mean
Shimmer-
>Ampl_TKEO_prc5
Shimmer-
>Ampl_TKEO_prc25
Shimmer-
>Ampl_TKEO_prc95
OQ->prc5_95
DFA
GNE->NSR_TKEO
GNE->NSR_SEO
VFER->mean
IMF->SNR_entropy
IMF->NSR_SEO
IMF->NSR_entropy
MFCC_2nd coef
MFCC_3rd coef
MFCC_6th coef
MFCC_7th coef
MFCC_8th coef
MFCC_10th coef
1st delta
2nd delta
5th delta-delta
Ea
Ed_8_coef
Ed_9_coef
Ed2_9_coef
entropy_shannon4_9_coef
entropy_log4_6_coef
entropy_log4_7_coef
entropy_log4_8_coef
det_TKEO_mean4_10_coef
det_TKEO_std4_3_coef
2. Tabu-
mRMR
8 Jitter->F0_TKEO_prc75
Shimmer-
>Ampl_TKEO_prc25
DFA
IMF->NSR_SEO
MFCC_2nd coef
MFCC_3rd coef
Ea
Ed_8_coef
Next, this reduced dataset is presented to the proposed approach that further optimizes the
dimensions of the data and finds an optimal set of features. At the end of this step a subset of
features is chosen for the next round. The optimal features by the Tabu and mRMR techniques
are listed in Table 4. It is worth noting that the number of features has remarkably reduced,
compared with mRMR technique, therefore less storage space is required for the execution of the
classification algorithms. In this phase we reduced the size of LSVT features from 309 to only 8
features. Table 5, shows the classification results of the first step. It can be observed that the
classification Accuracy using Tabu technique varies between 76% and 82% with the selected 37

10
feature set. In this step RF has achieved the results of 92.9%, 61.9%, 83%, 87.6% and 62.7%, in
terms of Sensitivity, Specificity, Precision, 𝑓-measure and MCC.
Table 5. Performance measures after first RF
Classifier Sensitivit
y
Specificit
y
Accuracy Precisio
n
f-
measure
MCC
RF 0.929 0.619 0.825 0.830 0.876 0.627
FURIA 0.893 0.619 0.802 0.824 0.857 0.561
RoughSet 0.821 0.643 0.762 0.821 0.821 0.464
AdaBoost 0.857 0.714 0.810 0.857 0.857 0.571
C4.5 0.881 0.632 0.803 0.841 0.860 0.542
In the next step, mRMR technique is applied on the selected features and the resulted feature set
is used as the inputs to the classifiers. In Table 6, we depict the performance of the classifiers
after applying the second step of reduction. From this table it is noticed that the highest Accuracy
rate is associated with RF classifier was 84.1% with 8 features. In Table 7, we can see the
comparative results of the classification performance of the second phase that deploy mRMR
algorithm to detect the most significant features. Clearly, we can observe that the mRMR helped
in reducing the dimension of features. Yet, this step did not improve the overall classification
performance.
Table 6. Performance measures of second stage RF
Classifier Sensitivit
y
Specificit
y
Accuracy Precisio
n
f-
measure
MCC
RF 0.929 0.667 0.841 0.848 0.886 0.662
FURIA 0.857 0.714 0.810 0.857 0.857 0.571
RoughSet 0.833 0.595 0.754 0.805 0.819 0.445
AdaBoost 0.833 0.643 0.770 0.824 0.828 0.482
C4.5 0.893 0.667 0.817 0.843 0.867 0.596
Tables 7 presents the last classification results of the suggested technique for mining the LSVT
dataset. The random resampling technique is applied on the second reduced dataset to increase
the samples of the minority class without replacement. The training set was resized at an over-
sampling rate of 100%, in order to balance the number of instances in the two classes. This step
should lead to make a more distinct and balanced dataset. We can observe that the highest
Accuracy rate is associated with RF classifier is 95.2% with 8 features. In this step RF has
achieved the results of 96.4%, 92.9%, 96.4%, 96.4% and 89.3%, in terms of Sensitivity,
Specificity, Precision, 𝑓-measure and MCC.
The results demonstrated that the reduced features are fairly sufficient to represent the dataset’s
class information. In terms of performance measures our proposed technique succeeded in
significantly improving the classification accuracy of the minority while the classification
accuracy of major class remains high. The outcomes from the suggested two-level attribute
selection techniques show better results compared to datasets which are not pre-processed and
also when these attribute selection techniques are used independently. It is clear now that our
developed method obtained promising classification results compared to the previously published
results in [7] which scored an average Accuracy of 90% using the original dataset.

11
Table 7. RF versus other machine learning techniques in predicting LSVT with resampling pre-processing
step
Classifier Sensitivi
ty
Specificity Accuracy Precisio
n
f-
measure
MCC
RF 0.964 0.929 0.952 0.964 0.964 0.893
FURIA 0.905 0.905 0.905 0.950 0.927 0.774
RoughSet 0.917 0.905 0.913 0.951 0.933 0.794
AdaBoost 0.881 0.738 0.833 0.871 0.876 0.627
C4.5 0.917 0.881 0.905 0.939 0.928 0.779
According to figure 2, the hybrid mRMR/Tabu technique has helped in investigating the LSVT
dataset. Starting from the raw dataset, we applied mRMR feature reduction technique to reduce
the features numbers. And in the second stage we applied Tabu Search technique to identify the
significant features. This helped in reducing the dataset from 309 to 8 features. Then these
features we fed into five classification algorithms. The RF algorithm performed scores were
notably higher in the two phases.
Figure 2. Comparison of proposed model based on accuracy
6. CONCLUSION
In this article, we have investigated a two-phase technique to improve the assessment of LSVT
dataset. We concluded that the proposed algorithm could improve the accuracy performance and
achieve promising results with fewer features that in the original dataset. The experiments have
shown that the Tabu and mRMR based RF classification strategy helped in reducing the feature
space. Whilst applying simple random sampling technique helped in adjusting the region area of
the minority class in favour of handling the existing imbalanced data property.
REFERENCES
[ 1 ] Aarli, J.A.; Dua, T.; Janca, A.; Muscetta, A. Neurological Disorders: Public Health Challenges.
World Health Organization: Geneva, Switzerland, 2006.
[ 2 ] De Lau, L.M.; Breteler, M.M. Epidemiology of parkinsonʼs disease. Lancet Neurol. 2006, 5, 525–
535.
[ 3 ] Ho, A. K., Iansek, R., Marigliani, C., Bradshaw, J. L., & Gates, S. (1998). Speech impairment in a
large sample of patients with Parkinson’s disease. Behavioural Neurology, 11, 131–138.

12
[ 4 ] Ciosa, K.J. and Mooree, G.W. (2002) Uniqueness of medical data mining,Artif. Intell. Med. 26 1–
24.
[ 5 ] Huang, S.H. Wulsin, L.R. Li, H. and Guo, J. (2009) Dimensionality reduction for knowledge
discovery in medical claims database: application to antidepressant medication utilization study,
Comput. Methods Programs Biomed. 93, 115–123.
[ 6 ] Luukka, P. (2011) Feature selection using fuzzy entropy measures with similarity classifier,
Expert Systems with Applications, Volume 38, Issue 4, 4600-4607, ISSN 0957-4174,
http://dx.doi.org/10.1016/j.eswa.2010.09.133.
[ 7 ] A. Tsanas, M.A. Little, C. Fox, L.O. Ramig: Objective automatic assessment of rehabilitative
speech treatment in Parkinson’s disease, IEEE Transactions on Neural Systems and Rehabilitation
Engineering, Vol. 22, pp. 181-190, January 2014.
[ 8 ] M. A. Little, P. E. McSharry, E. J. Hunter, J. Spielman, and L. O. Ramig, “Suitability of
dysphonia measurements for telemonitoring of Parkinson’s disease,” IEEE Transactions on
Biomedical Engineering, vol. .56, no. 4, pp. 1015-1022, 2009.
[ 9 ] Das, R. (2010). A comparison of multiple classification methods for diagnosis of Parkinson
disease. Expert Systems with Applications, 37, 1568–1572.
[ 10 ] Shahbaba, B., & Neal, R. (2009). Nonlinear models using Dirichlet process mixtures. The Journal
of Machine Learning Research, 10, 1829–1850.
[ 11 ] Sakar, C. Okan, and Olcay Kursun. "Telediagnosis of Parkinson’s disease using measurements of
dysphonia." Journal of medical systems 34.4 (2010): 591-599.
[ 12 ] Psorakis, I., Damoulas, T., & Girolami, M. A. (2010). Multiclass relevance vector machines:
sparsity and accuracy. Neural Networks, IEEE Transactions on, 21, 1588–1598.
[ 13 ] Guo, P. F., Bhattacharya, P., & Kharma, N. (2010). Advances in detecting Parkinson’s disease.
Medical Biometrics, 306–314.
[ 14 ] Luukka, P. (2011). Feature selection using fuzzy entropy measures with similarity classifier.
Expert Systems with Applications, 38, 4600–4607.
[ 15 ] Li, D. C., Liu, C. W., & Hu, S. C. (2011). A fuzzy-based data transformation for feature extraction
to increase classification performance with small medical data sets. Artificial Intelligence in
Medicine, 52, 45–52.
[ 16 ] Ozcift, A., & Gulten, A. (2011). Classifier ensemble construction with rotation forest to improve
medical diagnosis performance of machine learning algorithms. Comput Methods Programs
Biomed, 104, 443–451.
[ 17 ] AStröm, F., & Koker, R. (2011). A parallel neural network approach to prediction of Parkinson’s
Disease. Expert Systems with Applications, 38, 12470–12474.
[ 18 ] Spadoto, A. A., Guido, R. C., Carnevali, F. L., Pagnin, A. F., Falcao, A. X., & Papa, J. P. (2011).
Improving Parkinson’s disease identification through evolutionarybased feature selection. In
Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of
the IEEE (pp. 7857–7860).
[ 19 ] Sriram, TarigoppulaV.S., Rao, M.Venkateswara, Narayana, G.V.Satya & Kaladhar, D.S.V.G.K.
(2015). Diagnosis of Parkinson Disease Using Machine Learning and Data Mining Systems from
Voice Dataset. In Proceedings of the 3rd International Conference on Frontiers of Intelligent
Computing: Theory and Applications (FICTA) 2014. (pp. 151-157).

13
[ 20 ] Chuan Xu, Jiajun Chen, Xia Xu, Yingyu Zhang, and Jia Li, “Potential Therapeutic Drugs for
Parkinson’s Disease Based on Data Mining and Bioinformatics Analysis,” Parkinson’s Disease,
vol. 2018, Article ID 3464578, 8 pages, 2018. https://doi.org/10.1155/2018/3464578.
[ 21 ] Marras C, Beck JC, Bower JH, Roberts E, Ritz B, Ross GW, Abbott RD, Savica R, Van Den
Eeden SK, Willis AW, Tanner CM; Parkinson’s Foundation P4 Group. Prevalence of Parkinson's
disease across North America. NPJ Parkinsons Dis. 2018 Jul 10;4:21. doi: 10.1038/s41531-018-
0058-0. PMID: 30003140; PMCID: PMC6039505.
[ 22 ] C. Ding and H. Peng, “Minimum redundancy feature selection from microarray gene expression
data,” Journal of Bioinformatics and Computational Biology, vol. 3, no. 2, pp. 185–205, 2005.
[ 23 ] H. Peng, F. Long, and C. Ding, “Feature selection based on mutual information criteria of max-
dependency, max- relevance, and min-redundancy,” IEEE Transactions on Pattern Analysis and
Machine Intelligence, vol. 27, no. 8, pp. 1226–1238, 2005.
[ 24 ] El A, Aouatif A, El A, Driss O (2011) A two-stage gene selection scheme utilizing MRMR filter
and GA wrapper. Knowl Inf Syst 26:487–500
[ 25 ] Li A, Hu L, Niu S, Cai Y, Chou K (2012) Predict and analyze Snitrosylation modification sites
with the mRMR and IFS approaches. J Proteom 75:1654–1665
[ 26 ] F. Glover, “Tabu search—part I,” ORSA Journal on Computing, vol. 1, no. 3, pp. 190–206, 1989.
[ 27 ] F. Glover, “Tabu search—part II,” ORSA Journal on Computing, vol. 2, no. 1, pp. 4–32, 1990.
[ 28 ] Tahir, M.A., Bouridane, A., Kurugollu, F., Amira, A., 2004a. Feature Selection using Tabu Search
for Improving the Classification Rate of Prostate Needle Biopsies. In: Proc. 17th International
Conference on Pattern Recognition (ICPR 2004), Cambridge, UK.
[ 29 ] Tahir, M.A., Bouridane, A., Kurugollu, F., 2004b. Simulataneous Feature Selection and
Weigthing for Nearest Neighbor Using Tabu Search. In: Lecture Notes in Computer Science
(LNCS 3177), 5th International Conference on Intelligent Data Engineering and Automated
Learning (IDEAL 2004), Exeter, UK.
[ 30 ] Korycinski, D., Crawford, M., Barnes, J.W, and Ghosh, J., 2003. Adaptive feature selection for
hyperspectral data analysis using a binary hierarchical classifier and Tabu Search. In: Proceedings
of the IEEE International Geoscience and Remote Sensing Symposium, IGARSS.
[ 31 ] Sait, S.M., Youssef, H., 1999. General iterative algorithms for combinatorial optimization. IEEE
Computer Society.
[ 32 ] G. Weiss and F. Provost, "Learning when Training Data are Costly: The Effect of Class
Distribution on Tree Induction," J. Artificial Intelligence Research, vol.19,315-354,2003.
[ 33 ] Park, B.-H., Ostrouchov, G., Samatova, N.F., Geist, A.: Reservoir-based random sampling with
replacement from data stream. In: SDM 2004, 492-496, (2004)
[ 34 ] Estabrooks, A., Jo, T. & Japkowicz, N., 2004. A Multiple Resampling Method for Learning from
Imbalanced Data Sets. Computational Intelligence, 20(1), pp.18– 36.
[ 35 ] Yang, Y.Y. et al., 2011. Adaptive neural-fuzzy inference system for classification of rail quality
data with bootstrapping-based over-sampling. IEEE International Conference on Fuzzy Systems
(FUZZ-IEEE 2011), pp.2205–2212.
[ 36 ] Mitra SK and Pathak PK. The nature of simple random sampling. Ann. Statist., 1984, 12:1536-
1542.

14
[ 37 ] Ramig L, Pawlas A, Countryman S. The Lee Silverman voice treatment (LSVT®): a practical
guide to treating the voice and speech disorders in Parkinson disease. Iowa City, IA: National
Center for Voice and Speech; 1995.
[ 38 ] F. Provost, T. Fawcett, and R. Kohavi. The case against accuracy estimation for comparing
induction algorithms, Proceedings of the Fifteenth International Conference on Machine Learning,
San Francisco, CA: Morgan Kaufmann, 445-453, 1998.
[ 39 ] M.A. Maloof, Learning when data sets are Imbalanced and when costs are unequal and unknown,
ICML-2003 Workshop on Learning from Imbalanced Data Sets II, 2003.
[ 40 ] Smialowski P, Doose G, Torkler P, Kaufmann S, Frishman D: PROSO II-a new method for
protein solubility prediction. FEBS J 2012, 279(12):2192-2200.
[ 41 ] Oshiro TM, Perez PS, Baranauskas JA (2012) How many trees in a Random Forest? In: Machine
learning and data mining in pattern recognition. Springer, Berlin, pp 154–168 .

A MODIFIED MAXIMUM RELEVANCE MINIMUM REDUNDANCY FEATURE SELECTION METHOD BASED ON TABU SEARCH FOR PARKINSON’S DISEASE MINING

More Related Content

What's hot (17)

Similar to A MODIFIED MAXIMUM RELEVANCE MINIMUM REDUNDANCY FEATURE SELECTION METHOD BASED ON TABU SEARCH FOR PARKINSON’S DISEASE MINING (20)

Recently uploaded (20)

A MODIFIED MAXIMUM RELEVANCE MINIMUM REDUNDANCY FEATURE SELECTION METHOD BASED ON TABU SEARCH FOR PARKINSON’S DISEASE MINING