• Login
  • Register
  • Search

Discussion on Text Classification Algorithm Based on Genetic Algorithm and Probability Theory

Ling Sun


This article mainly studies the text classification algorithm based on genetic algorithm and probability theory, and improves the speed and accuracy of text classification by using the related knowledge of genetic algorithm and probability theory. Preliminary assignment of feature items is carried out through TF algorithm, and then special non-substantial words are shielded. Using L-E operator for weighting calculation can make the better results converge faster. Using genetic algorithm, using crossover operator, mutation operator and establishing a suitable objective function, speed up the retrieval speed and improve the efficiency of obtaining the best results. Using a hybrid algorithm can eliminate the interference of synonyms and non-characteristics.


Genetic Algorithm; Probability Theory; Text Classification; Bayesian Formula

Full Text:



[1] Zheng G. The law of algorithm and the algorithm of law. China Law Review 2018; (02): 66-85.

[2] David Belot. Probability Graphic Model. Beijing: People’s Posts and Telecommunications Press; 2018.

[3] Li B. The practical role of machine learning. Beijing: People’s Posts and Telecommunications Press; 2017.

[4] Li T, Du F. Research on fuzzy portfolio selection model based on background risk. Yinchuan: Ningxia Sunshine Publishing House; 2016.

DOI: http://dx.doi.org/10.18686/ahe.v5i2.3348