From Backend Engineer to Data Scientist

This notebook will demo the most common data input methods in Pandas.

• read_csv
• read_excel
• read_html
• read_sql

The location problem has been the challenge for many businesses starts for a long time. In this post, we try to find a more realistic location for new business, by employing Machine Learning methodologies according to existing data of related businesses. First, we fetch all related business data via FourSqare, and we define the sufficiency and insufficiency of the business, then we utilize Isolation Forest in Python Libary scikit-learn to find out the anomaly location since market anomalies can be great opportunities for investors. We also visualized the result in a map with the popular geo lib Folium.

This is my assignment of final project of IBM online course Machine Learning with Python via Coursera.

In this notebook we will build a classifier to predict whether a loan case will be paid off or not. We will load a historical dataset from previous loan applications, clean the data, and apply different classification algorithm on the data. The results is reported as the accuracy of each classifier, using the following metrics when these are applicable.