Similarity Judgment of Civil Aviation Regulations Based on Doc2Vec Deep Learning Algorithm
Haoyang Zhang, Liang Zhou · 2019
In view of such problems as the difficulty in updating the clustering results of text sets in real time after clustering, and the difficulty in retrieving the regulations of specific content caused by the continuous promulgation of laws and regulations, this paper introduces the Doc2Vec deep learning algorithm to improve the ability of text retrieval and updating the text classification results. Text similarity calculation has great research value in the field of statistical machine translation. Doc2Vec deep learning algorithm can update the clustering result by comparing text similarity without re-implementing text clustering. This paper will use the civil aviation legal provisions as data samples, use Doc2Vec deep learning algorithm to calculate the similarity between different legal provisions, and apply the method to the classification of legal provisions and the update of text clustering results. The experimental results show that although the method does not achieve the ideal expectation of the similarity between the texts in the same field, it can still classify the text more accurately.