CPC H04M 3/5233 (2013.01) [G06Q 50/06 (2013.01); H04M 3/2218 (2013.01); H04M 2203/408 (2013.01)] | 8 Claims |
1. A method for smart gas call center feedback management, wherein the method is implemented based on a smart gas management platform, the method comprising:
receiving a call message of a target customer through a call center, and a content of the call message being related to a gas business;
converting the call message to a text message;
determining a service category corresponding to the text message; and
determining, based on the service category, the feedback mode; wherein the determining, based on the service category, the feedback mode comprises:
determining, based on one or more of a service category of a message of other customers, a feedback mode of the other customers, and a service category of a message of the target customer, the feedback mode of the target customer through a preset algorithm; wherein the preset algorithm comprises:
establishing a training data set, a sample of the training data set including the service category of the message of the other customers and the feedback mode of the other customers; finding N counts of messages of the customers with the shortest distance to the message of the target customer in the training data set, and if among the N counts of messages, the messages in a feedback mode A are the most, then the feedback mode of the target customer is A; wherein the distance is related to a service category similarity, a real-time pipeline network situation, a customer positioning similarity, and an emergency similarity of the call message, the emergency similarity being predicted using a prediction model, the prediction model being a machine learning model including a first embedding layer, a second embedding layer, and an emergency similarity prediction layer; wherein
an input of the first embedding layer includes a text message of the target customer and a service category of the target customer, and an output includes a target customer feature;
an input of the second embedding layer includes a text message of the other customers and a service category of the other customers, and an output includes other customer features; and
an input of the emergency similarity prediction layer includes the target customer feature, the other customer features, a real-time pipeline network situation, a customer positioning of the target customer, and customer positionings of the other customers, and an output includes an emergency similarity;
in response to the feedback mode being manual feedback, determining a target operator through the call center to feed back a call of the target customer; and
in response to the feedback mode being automatic feedback, determining a feedback content through the call center and sending the feedback content to the target customer.
|