
Bivariate Oblique Decision Tree Algorithms Based on Linear Discriminant Analysis
Author(s) -
B Liu,
Jié Song,
S R Jiao
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1651/1/012084
Subject(s) - interpretability , oblique case , decision tree , linear discriminant analysis , algorithm , bivariate analysis , mathematics , boundary (topology) , computer science , tree (set theory) , data mining , artificial intelligence , statistics , mathematical analysis , philosophy , linguistics
To solve the problem of the low efficiency of the traditional decision tree algorithm in dividing decision boundaries that are not parallel to the coordinate axis as well as the problem of the difficulties and the limited interpretability of some oblique decision tree algorithms in solving high-dimensional data covariance matrix, the present study proposed a series of bivariate oblique decision tree algorithm based on LDA. In the algorithms, LDA was used to determine the division boundary, and impurity and variance analysis were used as a criterion for the selection of split variables, thus ensuring the interpretability of this method while improving the efficiency of defining the decision boundaries that were not parallel to the coordinate axis. Experiments conducted in multiple data sets show that algorithms has achieved satisfactory results.