CS 229 Final Report: Predicting Insurance Claims in Brazil
نویسندگان
چکیده
Improving the accuracy of insurance claims benefits both customers and insurance companies. Incorrect predictions effectively raise insurance costs for safe drivers and lower costs for risky drivers, and can be costly to insurance companies. Better predictions increase car-ownership accessibility for safer drivers and allow car insurance companies to charge fair prices to all customers. Better predictions also lead to improved profits for insurance companies. The problem is as follows: given a series of unlabeled features collected by an insurance company about a customer, can we predict whether the customer will file an insurance claim during a period of interest? The input to our algorithm is set of 595,213 labeled records, one per customer. Each record consists of n = 57 features with unknown meaning and a label indicated whether the customer filed an insurance claim. We then use least squares ridge regression, least squares lasso regression, logistic regression, Naive Bayes, random forests, gradient boosting, onelayer perceptron, and two layer-perceptron to predict whether a customer filed an insurance claim.
منابع مشابه
CS 229 = = Final Project Report SPEECH & NOISE SEPARATION
In this course project I investigated machine learning approaches on separating speech signals from background noise. Keywords—MFCC, SVM, noise separation, source separation, spectrogram
متن کاملA Social Network Analysis Framework for Modeling Health Insurance Claims Data
Health insurance companies in Brazil have their data about claims organized having the view only for providers. In this way, they loose the physician view and how they share patients. Partnership between physicians can view as a fruitful work in most of the cases but sometimes this could be a problem for health insurance companies and patients, for example a recommendation to visit another phys...
متن کاملCS 229 Project Report: San Francisco Crime Classification
Different machine learning approaches were conceptualized and implemented for predicting the probabilities of crime categories for crimes reported in San Francisco. The crimes records used in the research are downloaded from a competition on Kaggle. A Bayesian model, a mixture of Guassians model (stratified and unstratified), and logistic regression are implemented. A satisfactory result was ac...
متن کامل