faithmaks/socialmedia-customer-service

This is an attempt to ticket customer service complaints.

Customer Service Problem Classification.

Project Objective

The purpose of this project is to classify the complaints from Safaricom_Care twitter handle according to the department that is expected to handle them and ticket the issues based on priorities.

Methods Used

• Inferential Statistics
• Machine Learning
• Data Visualization
• Modeling

Technologies Used

• Python
• Pandas, textblob, matplotlib, scikit-learn, TextBlob, NLTK

Project Description

• We sourced our data from @Safaricom_Care twitter handle (link of data source) and we are seeking to identify the department in which these tweets need to be handled by.
• For data visualization we used WordCloud to get the words with the highest frequency.

Needs of this project

• Data understanding
• Data preparation/ data cleaning
• Modeling
• Evaluation
• Conclusion
• Recommendations
• Writeup

Getting Started

• Clone the repository.(for help see this tutorial)
• Raw data is being kept here
• The libraries that we used are here

Featured Deliverables

Write-Up

Contributing members

• Team Lead - Antony Brian
• Arnold Kalage
• Betty Bett
• Faith Makokha
• Julia Karanja

Contacts

Feel free to contact the team lead Antony Brian with any questions or if you are interested in contributing.