Features selection by genetic algorithm optimization with k-nearest neighbour and learning ensemble to predict Parkinson disease

Abstract

<span lang="EN-US">Among the several ways followed for detecting Parkinson's disease, there is the one based on the speech signal, which is a symptom of this disease. In this paper focusing on the signal analysis, a data of voice records has been used. In these records, the patients were asked to utter vowels “a”, “o”, and “u”. Discrete wavelet transforms (DWT) applied to the speech signal to fetch the variable resolution that could hide the most important information about the patients. From the approximation a3 obtained by Daubechies wavelet at the scale 2 level 3, 21 features have been extracted: a <a name="_Hlk88480766"></a>linear predictive coding (LPC), energy, zero-crossing rate (ZCR), mel frequency cepstral coefficient (MFCC), and wavelet Shannon entropy. Then for the classification, the K-nearest neighbour (KNN) has been used. The KNN is a type of instance-based learning that can make a decision based on approximated local functions, besides the ensemble learning. However, through the learning process, the choice of the training features can have a significant impact on overall the process. So, here it stands out the role of the genetic algorithm (GA) to select the best training features that give the best accurate classification.</span>

Evaluation of machine learning algorithms in the early detection of Parkinson's disease: a comparative study

“…The model that achieved the best performance was KNN with 95% in the metrics of precision, accuracy, sensitivity, and F1 count. On the contrary, studies [36], [41], [42] obtained lower results than this study with the same model, achieving 88.33%, 91.18% and 88% accuracy, respectively. Differentiating mainly with the studies [36], [41], where they used a different data set than the one used in this study.…”

“…The results position the DT model as the best predictor with 0.9477 in accuracy. The study [41], focused on the early detection of Parkinson's disease through the analysis of voice signals. Recordings of patients pronouncing vowels, processed using wavelet transforms, are used.…”

Best-worst northern goshawk optimizer: a new stochastic optimization method

“…In (6) states that the proportion of the movement away from the worst agent is reversely proportional to the iteration. In (7) states that the first candidate is obtained by combing the agent's current solution with movements generated from ( 5) and (6). In (8) states that this first candidate replaces the agent's current solution only if this first candidate is better than the current one.…”

“…Many old algorithms are still popular. Genetic algorithm (GA) is still implemented in recent engineering studies, such as hydrogen liquefaction process [2], town logistics distribution systems [3], construction projects [4], spectrometry peak detection [5], plain text encryption [6], Parkinson disease prediction [7], and so on. GA is also utilized in optimization studies related to the financial sector, such as in-stock selection [8], credit rating assessment [9], and cryptocurrency price forecasting [10].…”

An experimental study of tomato viral leaf diseases detection using machine learning classification techniques

“…Sample of major diseases of tomato in India (NHB, India database) Several studies were done earlier that show the capability of technology using algorithms of machine learning to identify objects in various sectors like retail, human behavior, face recognition [3], facial expressions [4], handwriting recognition [5], intrusion detection [6], movie recommendation [7], and food segmentation [8]. The health sector is also using it to identify/detect/predict various diseases like diabetes prediction, cancer detection [9], [10], heart disease, skin problems, Parkison's disease identification [11], COVID-19 from chest X-rays [12], and many more. The crop disease classification can also be done efficiently and effectively using machine learning.…”