Hi there! My name is Jace Yang. This was my final project of Categorical Data Analysis when I major Applied Statistics at Central University of Finance and Economics(CUFE) in 2020 fall.
-
code contains:
- Data Processing: Everything before data goes into model. I combined the Exploration Data Analysis with the feature engineering process.
- Logistics model, SVM model, and Decision Tree model are applied.
- funcs contains commonly used function across the codes.
-
data contains:
- the original data downloaded at UCI Bank Marketing Data Set,
[Moro et al., 2011] S. Moro, R. Laureano and P. Cortez. Using Data Mining for Bank Direct Marketing: An Application of the CRISP-DM Methodology.
- and some in-progress data stored in RData format to communicate with my teammates.
Note: The Report is written in Chinese, but it contains some fancy charts I created, so you can still take a look. Also, I will elaborate on the may idea in the following README page.
-
Forecast the success rate of the telemarketing calls, given information of target customers and the campaign records.
-
Customer profiling for a effective marketing strategy.
-
Personal Information:
-
Marital, Y Vs Job/Education/Housing:
- Account Information
-
Economics
- Age vs job/education
- Transform education type into education year:
-
year features and monthly contact features from labelled
month:
-
weekday of last contact:
-
previous and poutcome are similar, so just keep one.
We used Random Forest, Pmm linear prediction,
mode/Average, and combined them.
We used Logistics, SVM and decision tree.
Results:
