Research on Digital Serialization Method of Program Code
Baozhong Liu, Jianbin Liu, Jingjing Zhao, Chunlan Jiao · 2019
There are a large number of knowledge patterns in the massive code, which can be automatically obtained through data mining technology. Program code is a special data format, which can not be directly used as input to data mining algorithm. Aiming at this problem, this paper proposes a digital serialization method of program code. By this method, the program code data is converted into a data format that can be mined and the serialized data retains the structure information of the program code. The code serialization data with structural information guarantees the accuracy of the mining results to a certain extent. The classic data mining algorithms such as PrefixSpan, Spade and Spam are used to test the validity of the data format converted by the digital serialization method.