Intention prediction method and intention prediction device for network management and control system

Abstract

The invention discloses an intention prediction method and an intention prediction device for a network management and control system. The method comprises the steps: inputting training data into a neural network model for training, obtaining a first prediction model, and obtaining the prediction data of a time unit corresponding to the test data through the first prediction model; if a sudden change area exists between the prediction data and the test data, querying a user operation of a next time unit corresponding to the sudden change area through a knowledge graph, and generating correction data based on the user operation of the next time unit; and optimizing the first prediction model through the test data and the correction data to obtain a second prediction model, and predicting the intention of the user in the target time unit by adopting the second prediction model. Starting from the content of the user data packet, key features related to the user service are recognized by using the deep packet detection technology to represent the user intention, and the basis of intention understanding is improved.

Description

technical field [0001] The invention belongs to the field of communication networks, and more particularly relates to an intent prediction method and an intent prediction device for a network management and control system. Background technique [0002] The 5G era brings huge challenges to network operation and maintenance. The complexity of the network increases suddenly, the network traffic increases explosively, and the network cost remains high. Intent is an object that represents the user's purpose and operation intention. Through translation, the intent can be expressed as a specific service or network configuration, including one or more associated processing procedures. The current operation and maintenance management and control system of the communication network mostly passively recognizes and translates the user's intentions, pays less attention to the user's usage habits, lacks awareness of analysis and active operation and maintenance, and has limited user exper...

AI Technical Summary

Problems solved by technology

[0002] The 5G era brings huge challenges to network operation and maintenance. The complexity of the network increases suddenly, the network traffic increases explosively, and the network cost remains high.

And the prediction effect based on the data with low correlation is poor, and the intention based on the understanding of the operation sequence cannot be predicted

On the other hand, the relationship between the user's future intentions and historical intentions shows different characteristics such as randomness, continuity, or relevance in different scenarios. The modeling of the intention itself is relatively one-sided, and this prediction method will cause the algorithm to Overfitting leads to prediction errors, which in turn reduces the accuracy of intent predictions

[0004] Moreover, manual operation and maintenance is highly involved, the configuration process is cumbersome, the configuration cycle is long, the efficiency is low, the error rate is high, and the operation and maintenance cost and pressure are greatly increased; the response to network faults and problems is relatively passive, the network perception is poor, and the lack of active operation and maintenance Ability, resulting in poor customer experience

Images