The Analysis of Fuzzy C-Means and Pearson Correlation Methods for Data Reduction in kNN Algorithm

2021 ◽  
Author(s):  
Cut Desy Arisandi ◽  
Muhammad Zarlis ◽  
Syahril Efendi
Keyword(s):  
Data Reduction ◽  
Pearson Correlation ◽  
Fuzzy C Means ◽  
Correlation Methods

scholarly journals Methods of Estimating Correlation Coefficients in the Presence of Influential Outlier(s)

2021 ◽  
Vol 4 (3) ◽  
pp. 157-185
Author(s):  
Etaga H.O. ◽  
Okoro I. ◽  
Aforka K.F. ◽  
Ngonadi L.O.
Keyword(s):  
Correlation Coefficient ◽  
Pearson Correlation ◽  
Correlation Coefficients ◽  
Pearson Correlation Coefficient ◽  
Spearman Correlation ◽  
Correlation Methods ◽  
Linear Relationships ◽  
Median Correlation ◽  
Standing Problem ◽  
Research Analysis

Correlation methods are indispensable in the study of the linear relationship between two variables. However, many researchers often adopt inappropriate correlation methods in the study of linear relationships which usually leads to unreliable results. Recurrently, most researchers ignorantly employ the Pearson method in a dataset that contained outliers, instead of more appropriate correlation methods such as Spearman, Kendall Tau, Median and Quadrant which might be suitable in the calculation of correlation coefficient in the presence of influential outliers. It is noted that the accuracy of estimation of correlation coefficients under outliers has been a long-standing problem for methodological researchers. This is due to low knowledge of correlation methods and their assumptions which have led to inappropriate application of correlation methods in research analysis. Five different methods of estimating correlation coefficients in the presence of influential outlier (contaminated data) were considered: Pearson Correlation Coefficient, Spearman Correlation Coefficient, Kendall Tau Correlation Coefficient, Median Correlation Coefficient and Quadrant Correlation Coefficient.

Reliability of the Minimum Masking Level as Outcome Variable in Tinnitus Clinical Research

2020 ◽  
Vol 29 (3) ◽  
pp. 429-435
Author(s):  
Patricia C. Mancini ◽  
Richard S. Tyler ◽  
Hyung Jin Jun ◽  
Tang-Chuan Wang ◽  
Helena Ji ◽  
...  
Keyword(s):  
Standard Deviation ◽  
High Frequency ◽  
Pearson Correlation ◽  
Correlation Coefficients ◽  
Outcome Variable ◽  
Reliable Estimate ◽  
Moderate Correlation ◽  
Trained Subjects ◽  
The Relationship ◽  
Minimum Intensity

Purpose The minimum masking level (MML) is the minimum intensity of a stimulus required to just totally mask the tinnitus. Treatments aimed at reducing the tinnitus itself should attempt to measure the magnitude of the tinnitus. The objective of this study was to evaluate the reliability of the MML. Method Sample consisted of 59 tinnitus patients who reported stable tinnitus. We obtained MML measures on two visits, separated by about 2–3 weeks. We used two noise types: speech-shaped noise and high-frequency emphasis noise. We also investigated the relationship between the MML and tinnitus loudness estimates and the Tinnitus Handicap Questionnaire (THQ). Results There were differences across the different noise types. The within-session standard deviation averaged across subjects varied between 1.3 and 1.8 dB. Across the two sessions, the Pearson correlation coefficients, range was r = .84. There was a weak relationship between the dB SL MML and loudness, and between the MML and the THQ. A moderate correlation ( r = .44) was found between the THQ and loudness estimates. Conclusions We conclude that the dB SL MML can be a reliable estimate of tinnitus magnitude, with expected standard deviations in trained subjects of about 1.5 dB. It appears that the dB SL MML and loudness estimates are not closely related.

The CLASSIC/CLIMB Data Reduction: The Math

2014 ◽  
Vol 69-70 ◽  
pp. 75-99
Author(s):  
T. ten Brummelaar
Keyword(s):  
Data Reduction
Sign in / Sign up

Export Citation Format

Share Document