As a story-teller of data, I enjoy sharing my findings and insights of data science, exploratory analysis, and machine learning. Below are articles published in my Medium.
RFM Analysis & Customer Churn Analysis for Hotel/Mall Enterprise in China
In the article, I put down the code I use to create RFM metrics and do the clustering with them. The simple churn analysis is also documented.
Customer Funnel Analysis for Online Retailer — Using Pivot Table and Clustering in Python
Analytics process of web analytics - customer funnels. I conduct some EDA to identify key converter channel and also do the cluster analysis to differentiate groups of traffic.
WEB SCARPING H1B Salary DATABASE II: STATES AND COMPANY ASPECTS
The second article about H1B Database exploration. I mainly focus on the states and companies which issue the most H1B visa.
WEB SCRAPING H1B SALARY DATABASE I: mETHODOLOGY AND OVERALL COMPARISON
The first article about H1B Database exploration. In the article, I elaborate my thinking process of analysis, and compare three positions in business analytics field in general.
DETECTING EARNING MANIPULATION AND FRAUD iii: On a real-world company
Using tools and models from article 1 and 2, I try to investigate into if a real-world pharmaceutical company is potential to be a earning manipulator.
Detecting earning manipulation and fraud II: m-score generator
M-Score is a common tool for detecting earning manipulation. In the article, I share the process and code I create the tool that automatically query and visualize M-Score for a specific company. (The article was curated and selected by Analytics Vidhya)
DETECTING EARNING MANIPULATION AND FRAUD I: USEFUL TOOLS
Introduction of some useful tools to detect earning manipulation, and align with some codes for application. (The article was curated and selected by Toward Data Science)
text analytics of jack ma’s retirement letter and ELON MUSK’S PRIVATIZATION LETTER
Findings from the analytics results of two letter to reveal the future development of two giant companies. (The article was curated and selected by Toward Data Science)