brunoyang0918 / disaster-response

Disaster Response

This purpose to build this natural language model is to increase the speed of classify text message to the right category.

Initialization

The below pacakges are required for running these script. sys, sklearn, re, ntlk, pandas, numpy, sqlalchemy

File Description

1. process_data.py: A data clean function, which takes in raw data and output the combined datasets used in the modeling process. A typical way to call this funciton is like "python process_data.py disaster_messages.csv disaster_categories.csv"

2. train_classifier.py: A modeling function, which use the cleaned dataset to create a text classification model. A typical way to call this funciton is like "python train_classifier.py DisasterResponse.db classifier.pkl"

3. run.py The start function, which can start a webpage to display the dataset visualization and key-in message for classification. A typical way call this funciton is like "python run.py"