DheerajKumar97 / Customer-Life-Time-Value-Prediction-Flask-Deployment--Heroku

The motive of the project is to predict the Customer LifeTime Value of a Four Wheeler Insurance Company and it is implemented by satisfying all MLR Assumptions. All the basic Exploratory Data Analysis and Data preprocessing, end to end Data Science life cycle has been implemented in this project.

CUSTOMER LIFE TIME VALUE PREDICTION

Heroku App Link : https://customerlifetimevaluepred.herokuapp.com/

DEPLOYMENT OUTPUT

Table Of Contents

• PROJECT GOAL
• Project Motivation
• Requirements Installation
• File Section
• Technologies Used
• OLS REGRESSION MODEL OUTPUT
• License
• Sample EDA VISUALIZATIONS

PROJECT GOAL

This project is designed to predict the CUSTOMER LIFE TIME VALUE of four wheeler insurance company using Regression Analysis with Python, FLASK, HTML, SQL

A highly comprehensive analysis with all data cleaning, exploration, visualization, feature selection, model building, evaluation and MLR assumptions validity steps explained in detail.

Project Motivation

Every Organization runs with the goal to get a profit from their product and customers, most of the organization is workig hard without compromizing quality of products to help those organization business requirement, this project has been designed

Requirements Installation

The Code is written in Python 3.7. If you don't have Python installed you can find it here. If you are using a lower version of Python you can upgrade using the pip package, ensuring you have the latest version of pip. To install the required packages and libraries, run this command in the project directory after cloning the repository

pip install -r requirements.txt

File Section

In Customer Lifetime Value (Exploratory Data Analysis).py

1- Data Preprocessing and some Exploratory Data Analysis to understand the data

2- Data cleaning

In Customer Lifetime Value (Feature Engineering).py

1- Data preparation: Feature Engineering and Scaling

2- Feature Selection using RFE and Model Building

3- Regression Assumptions Validation and Outlier Removal

4- Rebuilding the Model Post Outlier Removal: Feature Selection & RFE

5- Removing Multicollinearity, Model Re-evaluation and Assumptions Validation

Details of Variables [Response Variable ==> Customer Life Time Value]

OLS REGRESSION MODEL OUTPUT

Data Points vs Fitted Line

Actual Points vs Fitted Points

Technologies Used

License

CopyRight 2020 DHEERAJ KUMAR

  https://www.gnu.org/licenses/gpl-3.0.en.html

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Sample EDA VISUALIZATIONS

#############################################################################################

Gender__vs__Policy Type

From this tabulation our insight will be

         (1) When compared to Male, Female gender has taken more Policies
         (2) In all three Policies, the most prefered or Taken Policy Type is Personal Auto Policy for both Male and Female

#############################################################################################

Marital Status__vs__Policy Type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are Married People

#############################################################################################

Employment Status__vs__Policy Type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all five categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are Employed People

#############################################################################################

Employed_People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are Employed and  Married People

#############################################################################################

UnEmployed_People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are UnEmployed and  Single People

#############################################################################################

Employed Male People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are Employed and Married Male People

#############################################################################################

Employed FeMale People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are Employed and Married FeMale People

#############################################################################################

UnEmployed Male People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are UnEmployed and Single Male People

#############################################################################################

UnEmployed FeMale People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all three categories is Personal Auto Policy
         (2) In all three Categories more Policy takers are UnEmployed and Single FeMale People

#############################################################################################

Married People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all five categories is Personal Auto Policy
         (2) In all five Categories more Policy takers are Married and Employed People

#############################################################################################

Married Male People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all five categories is Personal Auto Policy
         (2) In all five Categories more Policy takers are Married and Employed Male People

#############################################################################################

Married FeMale People vs Policy_type

From this tabulation our insight will be

         (1) Most prefered or Taken Policy Type for all five categories is Personal Auto Policy
         (2) In all five Categories more Policy takers are Married and Employed FeMale People

#############################################################################################

Scatter Plot Analysis

Histogram Analysis