https://github.com/akarshankapoor7/automated-complaint-triage-system-using-nlp-and-machine-learning
Automated Severity Classification of Forum Complaints for Resolution Teams - Emphasizes automation and the end goal for resolution teams.
https://github.com/akarshankapoor7/automated-complaint-triage-system-using-nlp-and-machine-learning
data-science datamining kmeans-clustering naive-bayes-classifier nlp tfidf-vectorizer
Last synced: about 2 months ago
JSON representation
Automated Severity Classification of Forum Complaints for Resolution Teams - Emphasizes automation and the end goal for resolution teams.
- Host: GitHub
- URL: https://github.com/akarshankapoor7/automated-complaint-triage-system-using-nlp-and-machine-learning
- Owner: akarshankapoor7
- Created: 2025-03-26T02:02:02.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2025-03-26T02:03:29.000Z (about 2 months ago)
- Last Synced: 2025-03-26T03:20:00.673Z (about 2 months ago)
- Topics: data-science, datamining, kmeans-clustering, naive-bayes-classifier, nlp, tfidf-vectorizer
- Language: Jupyter Notebook
- Homepage:
- Size: 25.4 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
The "Automated Complaint Triage System" is designed to analyze and prioritize customer complaints posted on a company forum. Using a predefined dataset of complaints with details such as customer ID, phone number, date posted, location, and email, the system employs natural language processing (NLP) and machine learning techniques to categorize issues into High, Medium, and Low severity levels. Initially, TF-IDF vectorization and K-means clustering assign tentative severity labels based on complaint text, which are then refined by a Naive Bayes classifier. The system evaluates its performance with classification metrics and a confusion matrix, enabling accurate predictions for new complaints. By providing prioritized action items with customer details and confidence scores, it streamlines the resolution process for support teams, enhancing efficiency in addressing customer concerns.