xingying99 Goto Github PK

blockchain-modreg

credit-card-payment-defaults-prediciton

This project aims to find the best classification model to predict credit card payment defaults using a given data set, which contains payment information of 30,000 credit card holders obtained from a bank in Taiwan. Each data sample is described by 23 feature attributes. The target feature (Attribute Y) to be predicted is a binary variable, 0 (= not default) or 1 (= default). We will be evaluating the models based on their ability to correctly predict clients that will and will not default on their payment next month. This is because it is equally important to predict both default and non-default cases as predicting the former would allow financial institutions to avoid financial losses to potential defaulters while predicting the latter would mean that credit applications will not be falsely rejected, compromising potential sales. Thus, the main metric for evaluation will be the average accuracy due to the skewness of the target variable and the desire to predict whether the person defaults or doesn't default correctly.

fraud-analytics-paper-reviews

portfolio-risk-managemet

predict-us-firm-bankruptcy

prediting-sales-of-video-games

question-and-answering

Capstone

spoon-knife

This repo is for demonstration purposes only.

yoursingapore