System, method, and computer program for preparing multi-feed of log data in an AI-managed communication system

A computer program product comprising computer executable instructions stored on a non-transitory computer readable medium that when executed by a processor instruct the processor to:

• obtain log data of a communication network, the log data comprising parameters associated with operation of network entities of the communication network and a time of reporting the parameters, and the log data further labeled with labels representing instances of at least one network situation of the communication network;
• obtain a plurality of artificial intelligence (AI) models, wherein each AI model of the plurality of AI models is configured to detect a respective classifier in the log data, the respective classifier preceding an instance of the at least one network situation;
• analyze the log data with a first AI model of the plurality of AI models to detect at least one occurrence of the respective classifier;
• compute a first confidence level for the first AI model, the confidence level representing at least one of:
• a first probability of the first AI model detecting the respective classifier;
• a second probability of the first AI model detecting the network situation;
• or
• a third probability of the network situation occurring after the detection of the respective classifier;
• form a reduced log data feed by eliminating at least one parameter from the log data;
• and
• compute a second confidence level for the first AI model by repeating the analysis of the log data, the computation of the first confidence level, and the formation of the reduced log data feed until at least one:
• the second confidence level is lower than the first confidence level;
• or
• the second confidence level reaches a threshold.

A device, comprising:

• a non-transitory memory storing instructions;
• and
• one or more processors in communication with the non-transitory memory, wherein the one or more processors execute the instructions to:
• obtain log data of a communication network, the log data comprising parameters associated with operation of network entities of the communication network and a time of reporting the parameters, and the log data further labeled with labels representing instances of at least one network situation of the communication network;
• obtain a plurality of artificial intelligence (AI) models, wherein each AI model of the plurality of AI models is configured to detect a respective classifier in the log data, the respective classifier preceding an instance of the at least one network situation;
• analyze the log data with a first AI model of the plurality of AI models to detect at least one occurrence of the respective classifier;
• compute a first confidence level for the first AI model, the confidence level representing at least one of:
• a first probability of the first AI model detecting the respective classifier;
• a second probability of the first AI model detecting the network situation;
• or
• a third probability of the network situation occurring after the detection of the respective classifier;
• form a reduced log data feed by eliminating at least one parameter from the log data;
• and
• compute a second confidence level for the first AI model by repeating the analysis of the log data, the computation of the first confidence level, and formation of the reduced log data feed until at least one:
• the second confidence level is lower than the first confidence level;
• or
• the second confidence level reaches a threshold.