Open AccessSignificance of Gender in Badminton Lunge Classification
Author(s) -
J J J Lee,
Wei Ping Loh,
Z W Ho
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/530/1/012057
Subject(s) - motion (physics) , computer science , statistical analysis , pattern recognition (psychology) , artificial intelligence , psychology , physical medicine and rehabilitation , mathematics , medicine , statistics
Most badminton lunge studies considered only the male players. No study has investigated the effect of gender on the badminton lunge movement. Past researches presented the statistical analysis approach on badminton lunge motion study. However, no study has classified lunge motion specifically by gender segregation. This paper investigates the gender effects in distinguishing the five lunge directions; center-forward (CF), left-forward (LF), right-forward (RF), left-lateral (L) and right-lateral (R) lunges using classification approach. The case study involved video captures of five-male and six-female university-level players lunging in the badminton 21-point singles. A total of 23 attributes of 10894 instances were extracted to classify lunges into CF, LF, RF, L and R lunges using 21 classification algorithms of the WEKA tool. Lunge direction classifications were performed on unsegregated and segregated genders datasets. Lunge classification on segregated gender datasets showed improvement over the unsegregated gender on 17 out of 21 classification algorithms. The notable improvements were on Trees-Hoeffding Tree (+8.67% on male, +3.08% on female) and Functions-Multilayer Perceptron algorithms (+4.35% on male, +9.37% on female). The study shows that gender segregation improves the lunge classification accuracies.