Research Article

Psychological Mechanism and Exercise Intervention of College Students’ Problematic Internet Use Based on IoT Technology

Table 2

Comparison of decision tree algorithms.

Decision tree algorithmAnalysis of algorithmsUse analytics

KNN methodThe K-nearest neighbor method is only related to a very small number of adjacent samples in the category decision, which can better avoid the problem of sample imbalance, but the amount of calculation is large.It is suitable for the sample set to be divided with more overlapping or overlapping class domains.

Bayes methodA pattern classification method with known prior probabilities and class conditional probabilities. It is necessary to obtain the probability distribution of the category population and the probability distribution function (or density function) of various samplesThe samples are required to be independent of each other, and the samples are large enough

Reverse KNN methodIt can reduce the computational complexity of KNN algorithm and improve the efficiency of classificationIt is suitable for automatic classification of the class domain with a relatively large sample size. If the sample size is small, it is easy to cause misclassification.

ID3 algorithm, C4.5, C5.0, and CART (classification regression tree)Using divide and conquer strategy, ID3 algorithm is the representative of each algorithm, and other algorithms are improved based on its principle, and use information gain to find the attribute field with the most information in the database to establish the node of the decision tree.It has the characteristics of simple description and fast classification speed, and is suitable for large-scale data processing.