Intelligent virtual customer service system

Abstract

The invention provides an intelligent virtual customer service system. The system is characterized by including: a basic database, which stores common customer service questions and answers; a keywordidentification module, which carries out word cutting on the common customer service questions and answers in the basic database, and gives single-node one-question-one-answer suggested-matching answers by fuzzy matching; and a process service module, which realizes question and answer interaction with a customer, extracts a corresponding answer from the basic database according to a question ofthe customer, and relocates a location of a question and answer node in context. According to the intelligent virtual customer service system of the invention, problems of high costs, difficult management, high losing rates and insufficient work times of existing manual customer services are solved, work efficiency of customer services is greatly improved, and a series of customer service work which is consulting, business transaction, customer complaint solving and the like and originally needs a large amount of labor for completion is completed in a real sense through humanized guiding-typecommunication process setting.

Description

technical field [0001] The invention relates to an intelligent virtual customer service system, which belongs to the field of intelligent customer service. Background technique [0002] Many studies have shown that in order to be able to learn complex functions representing high-level abstract concepts and solve artificial intelligence-related tasks such as target recognition, speech perception, and language understanding, deep learning needs to be introduced. The deep learning architecture consists of multiple layers of nonlinear computing units. The output of each lower layer is used as the input of the higher layer. It can learn effective feature representations from a large amount of input data, and the learned high-level representations contain many structures of the input data. Information, is a good way to extract representations from data, and can be used in certain problems such as classification, regression, and information retrieval. [0003] The concept of deep ...

AI Technical Summary

Problems solved by technology

[0006] 1. Since massive high-dimensional data requires a large-scale model to match it, the model and data can only be distributed and stored on a large number of points

[0007] 2. Despite the massive amount of data, due to the sparsity of the data, overfitting is still a problem that needs to be vigilant at all times

[0013] 2) Deep learning is not a black box system

[0021] 1. It is necessary to continuously set keywords, and the workload of customers is huge and heavy

[0022] 2. Need the customer team to cooperate with the system work and even the support of high-tech JAVA engineers

[0023] 3. Slow deployment, non-intelligent system, completed by traditional code, complex change process, 6-8 months on-line, heavy configuration system

[0024] 4. Unable to standardize, high cost, unable to provide services for small and medium-sized enterprises, only simple FAQ search answers, unable to carry out logical process design, no contextual connection, blunt experience, unable to truly solve user needs

[0025] 5. Unable to achieve a unified platform, each channel needs to establish a single independent robot and independent database

Images