Efficient learning of email similarities for customer support

2019

One way to increase customer satisfaction is efficient and consistent customer email support. In this contribution we investigate the use of dimensionality reduction, metric learning and classification methods to predict answer templates that can be used by an employee or retrieve historic conversations with potential suitable answers given an email query. The strategies are tested on email data and the publicly available Reuters data. We conclude that prototype-based metric learning is fast to train and the parameters provide a compressed representation of the database enabling efficient content based retrieval. Furthermore, learning customer email embeddings based on the similarity of employee answers is a promising direction for computer aided customer support.

@inproceedings{Bakker2018,
title = {Efficient learning of email similarities for customer support},
author = {Bakker, Jelle and Bunte, Kerstin},
booktitle = {Proc. of the  27th "European Symposium on Artificial Neural Networks (ESANN)},
pages = {119--124},
year = {2019},
publisher = {D-facto Publications},
editor = {M. Verleysen},
abstract = {One way to increase customer satisfaction is efficient and consistent customer email support.  In this contribution we investigate the use of dimensionality reduction, metric learning and classification methods to predict answer templates that can be used by an employee or retrieve historic conversations with potential suitable answers given an email query. The strategies are tested on email data and the publicly available Reuters data.  We conclude that prototype-based metric learning is fast to train and the parameters provide a compressed representation of the database enabling efficient content based retrieval.  Furthermore, learning customer email embeddings based on the similarity of employee answers is a promising direction for computer aided customer support.},
}