Support ticket classification

Many support tickets can be either automated or sent an automatic reply, other tickets will simply be routed to a different department. Automatically classifying these tickets and handling without needing to be triaged by first line support can improve response time and customer satisfaction and decrease the number of employees engaged in first line support.

Overview

Automatically classify support tickets into the appropriate group for processing.

TARGET

Class the support ticket is in.

Challenge

The primary information will be in the support database and contained on the support ticket, however understanding if this is a repeated request from within the support ticket database can improve customer experience when customers are having trouble with support. Adding information about the products the customer has purchased and their value to the company may allow for better classification and routing of tickets. All this data needs to be cleaned, joined and transformed into valuable ML features before going into model training. This pre-modeling prep process can be frustrating and time consuming. We are here to help.

Modeling techniques and libraries

Machine learning analysis

Build machine learning models to predict the target as a function of the independent variables. Use model interpretability packages to evaluate the impact of the independent variables on the prediction.

Packages:
  • Sklearn
  • ELI5
  • LIME
  • SHAP
  • NLTK
  • CoreNLP
  • Gensim
  • spaCy
  • Polyglot
  • Pattern

Data features

# of Support Tickets
Customer Support DB
Data Type
Continuous
Target
No
Yes
Avg Resolution Time per Ticket
Customer Support DB
Data Type
Continuous
Target
No
Yes
Common Words from Support Interactions
Customer Support DB
Data Type
Categorical
Target
No
Yes
Customer Lifetime Value
CRM
Data Type
Continuous
Target
No
Yes
New Support Tickets per Day
Customer Support DB
Data Type
Continuous
Target
No
Yes
New Support Tickets per User
Customer Support DB
Data Type
Continuous
Target
No
Yes

Related accelerators

No items found.

Get your data science on.