Kaggle_Competitions

It's about my analysis on large and real life problem based competitions @kaggle and Applied Data analysis, machine learning and Deep learning techniques to build necessary model. Follow me on @Kaggle : https://www.kaggle.com/harshkothari21

1. House Price

Predict House-Price on DataSet that conatins 81 features and 3000 Rows(1460 for Train and 1459 for Test).

Skills Applied:

• Data Cleaning
• EDA
• Feature Selection
• Feature Engineering
• Random Forest Model and XgBoost
• Hyperparameter tuning
• Deep Learning using Keras

2. Titanic

Classification problem to predict weather the person survived or not during the famous titanic Accident.

Skills Applied:

• Data Cleaning
• EDA
• Feature Engineering
• Scaling
• Cross Validation
• Hyperparameter tuning
• Logistic Regression | SVM | SVC | KNN | Decision Tree |Random Forest Model

3.Career Village

This competition contains dataset from https://www.careervillage.org/ , We need to be able to send the right questions to the right volunteers and get the insights of the data.

Skills Appied:

• Data Cleaning
• EDA
• Data Visualization
• Build WordCloud

4.Don't Overfit the Model

A binary classification task to train a model with 300 features and only 250 training examples and 79times more samples on test data.