Method for detecting software self-acceptance type technology debt

The invention discloses a method for detecting software self-acceptance type technology debt. The method comprises the following steps: firstly, acquiring and processing a data set; constructing a self-acknowledged technology debt detection model comprising three parallel base classifiers; wherein the three parallel base classifiers are a CNN model, a CNN-LSTM hybrid model and a DPCNN model; inputting the word vector matrix into three base classifiers, wherein each base classifier outputs the probability that the annotation data belongs to the self-acknowledgement type technology debt; fusing the classification results output by the three base classifiers to acquire the final probability that the annotation data belong to the self-acknowledged technology debt; and finally, judging the size relationship between the probability and the classification threshold, and outputting a result for detecting whether the annotation data is the self-acknowledged technology debt or not. The problem that the misjudgment rate