A method and system for improving the accuracy of predicting outbound calls based on number attribution

Abstract

The invention discloses a method and system for increasing the call-out prediction accuracy rate on basis of number attribution. The method comprises the steps that a prediction algorithm is conducted; a number group is extracted from time zone groups divided from an actual time zone where numbers needing to be called out at a current batch are located according to the numbers needing to be called out at the current batch; call-out is initiated according to the number group extracted from the time zone groups divided from the actual time zone where the numbers needing to be called out at a current batch are located. According to the method for increasing the call-out prediction accuracy rate on the basis of number attribution, it is guaranteed that the number proportion of all areas can keep consistent in the predicted call-out numbers initiated at all batches, the influences of different area number fluctuations on a prediction result are shielded, and then the call-out prediction accuracy rate is increased; due to the fact that the number extracting process is conducted after the total call-out number of the current batch is calculated, the method can be independent of the algorithm, and therefore the method can be suitable for different prediction algorithms.

Description

technical field [0001] The present invention relates to the technical field of predicted outbound calls, in particular to a method and system for improving the accuracy of predicted outbound calls based on number attribution. Background technique [0002] Mobile phone numbers in China, including those of China Mobile, China Unicom, and China Telecom, have a fixed regional distribution, and this region is where they belong. A mobile phone number, its attribution is the same, no matter how many people have registered and canceled the number. Just like the area code of a city in China, it remains unchanged. [0003] Predictive outbound calls are a type of batch outbound calls in the call center. This outbound dialing method can effectively filter out invalid numbers, assign connected calls to agents, and agents can communicate with customers after off-hook, greatly improving work efficiency. [0004] In the existing predictive outbound call process, calls to the same number w...

AI Technical Summary

Problems solved by technology

[0004] In the existing predictive outbound call process, calls to the same number will have different answering rates at different times of the day. For example, customers are more willing to answer calls from the call center during non-working hours; due to China's vast territory, the longitude spans Large, the work and rest time of cities distributed on various longitudes are quite different; if the numbers are not distinguished during the process of predicting outbound calls, in each batch of outbound numbers, the proportion of numbers belonging to each area is inconsistent, resulting in The connection rate of numbers initiated by each batch is different, which affects the accuracy of predicting outbound calls

Images