Belief Decision Tree

Cerebral Palsy Classification Using Heuristics and Belief Decision Tree ... 

Keywords— Classification, Discrimination, Cerebral Palsy, Belief Decision Tree, Heuristics, Diplegia, Hemiplegia. 

Belief decision tree method is a classification method which has been developed recently based on belief functions [13], [14]. This method deals with a universal formalism for classifying uncertain classes with belief level. Furthermore, its promising applications are welcome to any field of sciences when performing uncertain discrimination [15]. 

The present study aims to develop an uncertain classification process of cerebral palsy using a heuristics-based belief assignment method and belief decision tree method. This helps clinicians to classify two main types of cerebral palsy (diplegia and hemiplegia) with belief label automatically based on gait kinematics, kinetics, and EMG parameters. 

Belief decision tree is the combination of standard decision tree and belief fimctions. Basic elements of a belief decision tree consisted of a decision node, an edge, and a leaf with uncertain label. Averaging approach was selected to define the attributes selection measures based on extended gain criterion of the information theoiy of Shannon. A mean value was used as partitioning strategy. The growth process of belief decision tree was stopped when there are no more attributes to be tested [13]. 

The mass distributions of all subjects based on the difference of ground reactions forces of heel strike and toe off phases (Table 2) and EMG-based average rectified voltage were calculated. The mass distribution of the combination of the difference of ground reactions forees of heel strike and toe off phases and EMG-based average rectified voltage was carried out as input data in order to develop the belief decision tree of the cerebral palsy. 

The belief decision tree of cerebral palsy showed an impact of the ground reactions forces and EMG-based average rectified voltage for classifying two types of cerebral palsy such as spastic diplegia and hemiplegia. The combination of belief functions and standard decision tree permitted to develop a universal formalism in case of uncertain classification. 

As we might expeet by this stage in the book, most of the usual classification/regression algorithms have been applied to the duration predietion problem. These include decision trees [372], neural networks for phone prediction [109], [157], genetic algorithms [319] and Bayesian belief networks [182], Comparative studies of deeision trees and neural networks found little difference in accuracy between either approaeh, [473], [187], [72],... 

---