In view of the characteristics of various types of information system integrated hardware and software systems, complex network topology, complex causes of fault alarm and uncertainty, this paper studies the communication fault maintenance decision-making based on the inverse symmetry algorithm. Based on the principle of the inverse symmetry algorithm, the modulation and demodulation process of information system is determined, the redundant system attributes in the information system are reduced by rough set, and the Bayesian network model with minimum diagnosis set is obtained by combining the prior knowledge in the operation process of the information system. The input and output of the network are the condition attributes and decision attributes of the decision table, respectively. Through the above process, the optimal fault diagnosis rules are established. After the communication fault is diagnosed, different communication fault maintenance decisions of the information system are used to eliminate the fault. The experimental results show that this method can effectively diagnose the communication fault of the information system and make effective maintenance decisions. The average accuracy of data transmission of the information system using this method is over 99.5% under different operating distances, which has better communication performance.