International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 05 | May 2022
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
Bank Customer Segmentation & Insurance Claim Prediction Yashi Rajput1, Prof. Vikash Singhal2, Manish Saraswat3, Vanshika Chitranshi 4, Mohd. Talib Khan 5, Shipra Shrivastava 6 1,3,4,5B-Tech
Student, Greater Noida Institute of Technology, Greater Noida, India Greater Noida Institute of Technology, Greater Noida, India 6Ass.Professor, Greater Noida Institute of Technology, Greater Noida, India 2Professor,
-------------------------------------------------------------------------------***-----------------------------------------------------------------------------1.1 CLUSTERING Abstract - This project will help a bank in segmenting their client and an Insurance company to study the claim Prediction pattern. This design is majorly grounded on data mining and its ways. It majorly focuses on clustering and Prediction using machine literacy and python libraries similar as NumPy, pandas, seaborn, and matplotlib. It also contributes to the enhancement of the pricing models. This helps the insurance company to be one step ahead of its contender. Carrying and acting on client data through the lens of segmentation can have a massive impact on marketing and deals, retention sweats, client service, and more.
A bank’s client segmentation (Clustering) approach can vary extensively and must be grounded on the association’s business model and precedences. Parts can be quantitative, similar as by age and gender, or they can be qualitative, similar as separation by values and interests. The maximum value is attained when banks combine both types of data to more understand the wants and requirements of their client parts, allowing them to offer the right product or service at the right time.
Key Words:
NumPy, Pandas, Seaborn, Matplotlib, Machine learning, Decision Tree, Random Forest, Binary logistic Regression.
A leading bank wants to develop a client segmentation to give promotional offers to its guests. They collected a sample that summarizes the conditioning of druggies during the once many months. We've to identify the parts grounded on credit card operation.
1. INTRODUCTION Client segmentation is the approach of dividing a large and different client base into lower groups of affiliated guests that are analogous in certain ways and applicable to the marketing of a bank’s products and services. Some introductory segmentation criteria include terrain, income, and spending habits. Through client segmentation, banks can get to know their guests on a core particular position and offer them more customized products and services.
1.2 CART-NF-ANN An Insurance Claim Prediction (Cart-NF) firm providing tour insurance was facing higher claim frequency. The management decided to collect data from the past few years. We will make a model that predicts the claim status and provides recommendations to management. Using CART, RF & ANN or comparing the models' performances in train and test sets.
Insurance companies are extensively interested in the Prediction of the future. Accurate Prediction gives a probability to drop fiscal loss for the company. The insurers use rather complex methodologies for this purpose. The major models are a decision tree, a arbitrary timber, a double logistic retrogression, and a support vector machine. A great number of different variables are under analysis in this case.
2. WORKING It is divided into two sub problems:
2.1 Bank Customer Segmentation Step 1: Doing all the initial steps, and exploratory data analysis (Univariate, Bi-variate, and multivariate analysis).
A bank’s client segmentation approach can vary extensively and must be grounded on the association’s business model and precedences. Parts can be quantitative, similar as by age and gender, or they can be qualitative, similar as separation by values and interests.
Step 2: Justifying that scaling is necessary for clustering in this case. Step 3: Applying hierarchical clustering to scaled data. Identifying the number of optimum clusters using Dendrogram and briefly describing them.
The maximum value is attained when banks combine both types of data to more understand the wants and requirements of their client parts, allowing them to offer the right product or service at the right time.
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 1327