Simulation Research on Mathematical Model of Network Node Matching Problem Based on Fuzzy Mathematics

Traditional matching methods rely on accurate mathematical models, but in the real world, the characteristics or attributes of network nodes are often fuzzy and difficult to describe with accurate mathematical language. This study explores the mathematical model of network node matching based on fuzzy mathematics to improve the flexibility and accuracy of matching. Firstly, the fuzzy sets of node characteristics in the network are defined, and based on these fuzzy sets, the fuzzy relationship of similarity between nodes is defined. By introducing cosine similarity measurement method, the matching rules based on fuzzy mathematics are constructed, and the matching threshold is set to screen out the matched node pairs. In this study, Python language and NumPy, Pandas and Matplotlib libraries are used for data processing and simulation experiments, and the effectiveness and accuracy of the model are verified by using Facebook social network data set as experimental data. The experimental results show that the model can achieve a better F1 score under an appropriate matching threshold, and achieve a balance between accuracy and recall. In addition, with the increase of the matching threshold, the accuracy of the model increases and the recall rate decreases, indicating that it is necessary to choose the appropriate matching threshold according to the actual needs to balance the performance. The model in this study can effectively deal with the fuzziness of node characteristics, and is suitable for the problem of network node matching with fuzzy characteristics, and provides a foundation for further optimizing the model algorithm and improving the computing performance in the future.