Predictive analysis of sewer sewage purification

Due to the sewer drainage, the sewer reservoir needs to be cleaned from time to time. The amount of scavenger required to purify the water needs to be based on the experience of the operators and is related to external uncontrollable factors such as weather. If the amount is too small, the purification effect will not be achieved, and if the amount is too much, the odor will occur. This project uses machine learning methods to predict the amount of purifier input amount. Regular meetings with clients to discuss project progress. Analyze client’s data using statistical methods, and report to the client. Use linear regression, decision tree and other machine learning methods to analyze variables, select variables, etc. I used clustering to extract unknown features. Write a web crawler to capture the public data of a third-party website for machine learning modeling. Write a simple predictive application to simplify the client’s business operations.

Tools: Python3, Matplotlib, Pandas, Scikit-learn, Jupyter Notebook, Git, Docker