To view the code with exploratery plotly graphs click on below link...
Loan approval prediction is traditional preoblem and given at various Hakathon on various platforms such as Kaggle, Analyticsvidhya. Training datset consists of Gender, Married, Dependents, Education, Self_Employed, ApplicantIncome, CoapplicantIncome, LoanAmount, Loan_Amount_Term, Credit_History, Property_Area and Loan_Status (Approved/Rejected). Using this data a Machine Learning model is to create such that next time it will predict that if Loan for particular application is to approve or not.
To understand the data and to get most possible information from the data Explorary Data Analysis is done as first part. Plotly, Matplotlib and Seaborn python libraries are used for data visualization. plotly 4.9.0 matplotlib 3.3.0 seaborn 0.10.0
To try most possible ways to get maximum accuracy number of Machine Learning Algorithms are tried with data with Parameter tunning. Also tried with Voting Ensemble Method. Aim was to check efficiency of classifiers on the given data.
Extream Gradient Boosting Random Forest Decision Tree Gradient Boosting Ada Boost Support Vector Machine Logistic Regression Linear Discriminant Analysis KNearest Neighbors Gaussian Naivey Bayes
Maximum accuracy achieved by Random Forest of 83.88%. Average accuracy achieved by all classifiers is 83.11%.
More improvement in terms of Data analysis , Data wrangling, Model selection will be tried to increase accuracy.
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)