The dataset of the project is about the admission status of high school graduates applying for universities. Mainly, this project is done by Exploratory Data Analysis (EDA). EDA is one of a significant stage in analyzing data. In this stage, the data must be understood in order to obtain a great insight from it. The steps to do the EDA are:
- Preliminary look at the data
- Data cleaning (missing value and duplicate rows handling)
- Data deep-dive understanding (statistical summary, univariate analysis, and multivariate analysis.)
Conclusions of this project are:
- The data does not have huge problems. There are only few NULL values and duplicated rows.
- The type of data and the maximum and minimum values in each column are appropriate.
- Most of the columns is symmetrically distributed and outliers are still normal.
- In the bar graph of research experience vs admit status, many students who have research experiments are admitted to the university.
- In correlation heatmap, there are 3 columns of gre score, toefl score, and gpa that are highly correlated and redundant. Therefore, choose 1 of them for modelling.
- Students who have high score in GPA, gre score, and toefl score are admitted in the university.
- Students with good motivation letter and recommendation strength are admitted in the university.