cyberquantic logo header
EN-language img
FR-language img
breadcrumbs icon
Retail

Customer Sentiment Analysis

Emotion-sensitive AI customer service

Emotion-sensitive AI customer service

For:
Customers using the customer service system
Goal:
Other
Problem addressed
To design an efficient solution for detecting customers sentiment and its
intensity, especially in situations for which there is a limited training dataset.
Scope of use case
Extracting sentiment and its intensity from customers input, and responding
with an appropriate attitude in order to improve the quality of customers
inquiry experience.
Description
JDs customer service representatives are expected to handle
millions of requests on a daily basis. Regular AI customer
service systems, which are online 24/7, are capable of
offering instant assistance, which frees up labour resources
to a large extent. However, it is quite challenging, if not
impossible, for those systems to interpret emotions from
customer input and respond in as friendly a manner as a
human being.
Against this background, based on a huge data set of
customer comments and the rich experience of Natural
Language Processing, our system can automatically detect
sentiments like "happy, angry, anxious," etc. Moreover, this
system can also detect the intensity of the customer's
sentiment. Furthermore, we adapt Convolutional Neural
Networks, a widely used technique in visual computing, to
interpret the semantic meaning of the customers
expression. It can improve the systems performance for
sentiment classification and intensity detection. Moreover,
with the adoption of transfer learning, the system can also be
applied to various types of data. To overcome the difficulty
of limited training data, we also use data augmentation
methods such as reverse translation and data noise to
increase the variability of training data.
Up to now, the system has reached 90 % recall and 74 %
accuracy for sentiment classification over seven categories.
The overall recall and accuracy for sentiment intensity are
also around 85 %. The system has increased customer
satisfaction by 57 %.
understand frame img
transfer learning
Interested in the same or similar project?
Submit a request and get a free project evaluation.