- Problem: Each week the Consumer Financial Protection Bureau sends thousands of consumer’s complaints about financial product and services to company for a response. Classify those consumer complaints into the product category it belongs to using the description of the complaint.
- Solution: The goal of this project is to classify the complaint into a specific product category. Since it has multiple categories, it becomes a multiclass classification that can be solved through many of the machine learning algorithms.
- Once the algorithm is in place, whenever there is a new complaint, we can easily categorize it nad can then be redirected to the concerned person. This will save a lot of time because we are minimizing the human interventions to decide whom this complaint should go to.
we're going to be using some real banking and finance consumer complaint data Samples. Go to the link and download the data: here
- Hit the green button and clone this repo.
- See Notebook: ConsumerComplaintClassification.ipynb is used to build multiclass classfication model for categorize complaint_narrative of customers to there product category based on which will be further communicated to the concern issue resolving team.