Outbound model construction method, outbound method and device and storage medium

Abstract

The embodiment of the invention discloses an outbound model construction method, an outbound method and device and a storage medium. The outbound model construction device in the application acquiresa client identifier and historical outbound time of a historical outbound client, and historical outbound times and historical connection times corresponding to the historical outbound time; the historical call-out time is classified into a preset time period type according to the time period of the historical call-out time; according to the historical call-out times and the historical call-through times, the time period call-through rates corresponding to the time period types are determined respectively; a sample set is constructed according to the customer identifier, the time period type and the time period call completing rate; and finally, an intelligent outbound model is constructed according to a preset model and the sample set. According to the scheme, the model is constructed only after the historical outbound time is changed into the time period type of the time period where the historical outbound time is located, the prediction accuracy of the intelligent outbound model can be improved, and then the call completing rate of intelligent outbound is increased.

Description

technical field [0001] The invention relates to the field of outbound calls, in particular to a method for constructing an outbound call model, an outbound call method, a device and a storage medium. Background technique [0002] With the development of e-commerce and the logistics industry, the volume of express shipments has grown rapidly. Logistics companies need to deal with the many abnormal situations of express shipments that follow. [0003] In order to reduce labor costs, the application of the outbound call system has become the choice of many companies. However, the existing outbound system only makes outbound calls to customers based on the customer's historical connection rate at each time point. For example, the system once failed to call the customer at 9:00 in the morning. When the module is training, it will only mark the failure of the outbound call at 9:00 am. The next time the outbound call is made at 10:00 am, it will not consider the record of the pre...

AI Technical Summary

Problems solved by technology

However, the existing outbound system only makes outbound calls to customers based on the customer's historical connection rate at each time point. For example, the system once failed to call the customer at 9 o'clock in the morning. When the module is training, it will only mark the outbound call failure at 9:00 a.m., and will not consider the record of the previous 9:00 a.m. outdial call failure when the next outbound call is made at 10:00 a.m. The time slot for answering calls, for example, I don’t want to answer these calls in the morning, but I will answer them at night, so the pass rate of Chinese and foreign calls is not high in the existing technology, and the logistics company hopes to predict more accurately whether the customer will be connected. Outbound calls, thereby reducing the number of outbound calls that will not be connected, and improving the completion rate of outbound calls

Images