Learning a Financial Institution's decision making approach.
If you are looking to skim over the project without going into too much detail, you can always access it through here.
The Canadian banking system continues to thrive despite constant changes in financial innovations and technological integrations. As a result, Banks and Financial Institutions are periodically being audited in order to satisfy OSFI's (Office of the Superintendent of Financial Institutions) stringent rules and regulations. However, as long as an audit involves Human interactions, then this creates an opportunity for Human errors to occur. For example, if a program is 100% accurate, Humans would still disagree 20% of the time. The reason for this is that Humans can make bias errors and opinion errors from our decision making process, while on the other hand, computers do not make the same bias errors and opinion errors, but instead make different errors which is not comparable to Human errors (ie. understanding jokes, sarcasm, etc). I will be attempting to create an "Internal Audit System" to avoid errors that Auditors can potentially make 20% of the time.
The dataset I will be using is directly from a Canadian Bank, Although we were given permission to showcase this project, however, we will not showcase any relevant information from the actual dataset for privacy protection.
This project was completed using Jupyter Notebook and Python with Pandas, NumPy, Matplotlib, Scikit-Learn and Eli5.
I was able to dive into real world data and apply it in a business context. I will continue to upload interesting topics in the future to enhance a Financial Institution's capabilities using Machine Learning and Deep Learning!