#CREDIT CARD VALIDATOR PYTHON PROJECT INSTALL#
#CREDIT CARD VALIDATOR PYTHON PROJECT SOFTWARE#
#CREDIT CARD VALIDATOR PYTHON PROJECT CODE#

The ‘Time’ feature comes from the bimodal distribution which can’t be transformed to gaussian-like, thus we will discard it. For example, you can try this technique on features ‘V26’, ‘V4’, ‘V1’ and see if you get improvements in our final score. This method works only for features from the gaussian distribution, thus if your data is not gaussian-like you can transform it to be with a method that I describe in this article. Therefore, this will be our motivation to do fraud detection with the multivariate Gaussian distribution. We see that almost every feature comes from the normal (or gaussian) distribution except from the ‘Time’ one. The blue line indicates the actual Gaussian distribution, while the red one our data’s probability density function.

#CREDIT CARD VALIDATOR PYTHON PROJECT CODE#

I encourage you to execute the above code snippet in your local machine to see the results better. Thus the next graph will be the data distribution one: fig, axs = plt.subplots(6, 5, squeeze=False) for i, ax in enumerate(axs.flatten()): ax.set_facecolor('xkcd:charcoal') ax.set_title(df.columns) sns.distplot(df.iloc, ax=ax, fit=norm, color='#DC143C', fit_kws=) ax.set_xlabel('') fig.tight_layout(h_pad=-1.5, w_pad=-1.5) plt.show() Until now, we did some nice data analysis but we didn’t do the most important part which is, learning “where our data came from”. There is also a little correlation between ‘Time’ and other features but we will see a more important reason to drop it in the aftermath of this article. The ‘Amount’ feature is slightly correlated with the other features as well, meaning that it can partially be calculated by them so we can try dropping it (from my tests, this gave a nice amount of improvement to the final score). It seems that we don’t have highly correlated features although there is a slightly negative correlation between ‘V2’ and ‘Amount’ (label not shown in the above heatmap but it is near the ‘Class’ feature) features. Let’s plot how balanced our dataset is: # plot normal and fraud count_classes = pd.value_counts(df, sort=True) count_ot(kind='bar', rot=0) plt.title('Distributed Transactions') plt.xticks(range(2), ) plt.xlabel('Class') plt.ylabel('Frequency') plt.show() There are no missing values so we can proceed with the visualizations to better understand our data. Now we will read the dataset and check if there is any missing value: df = pd.read_csv('creditcardfraud/creditcard.csv') # missing values print('missing values:', df.isnull().values.any()) Initially create a ‘main.py’ file and import the following: import numpy as np import pandas as pd import sklearn from scipy.stats import norm from scipy.stats import multivariate_normal from sklearn.preprocessing import MinMaxScaler import matplotlib.pyplot as plt import seaborn as sns

#CREDIT CARD VALIDATOR PYTHON PROJECT INSTALL#

Initially, we will install the required packages so grub the ‘requirements.txt’ file from my GitHub repo, create a virtual environment install them: pip install -r requirements.txt ‘Amount’ is the price of each transaction, ‘Time’ is the > and ‘Class’ represents a fraud transaction when its value equals 1 and a valid transaction when its value equals 0. The dataset consists of credit card transactions whose features are the product of PCA analysis and thus we don’t know what they represent except from the ‘Amount’, ‘Time’ and ‘Class’ ones. You can find the code and the dataset in my github repo but I strongly encourage you to build it along with the instructions of this article. In this article, I will present a way to detect if someone bypasses the security walls and makes an illegal transaction. For example, if you follow the django framework release notes, you will see that there are many bug fixes among versions, thus a company should not rely only on its engineering skills. With careful designing most (if not all) of the problems can be eliminated but even the framework which was used to create the server is not perfect. Those people are designing the company’s servers in a way that the client has no control over critical transaction parts such as payment amount.

#CREDIT CARD VALIDATOR PYTHON PROJECT SOFTWARE#

It seems that the digital age brought some highly useful features which changed the way that both companies and consumers interact with each other but with one cost… Companies need to hire skilled software engineers and penetration testers to make sure that all the transactions are legal and non-fraudulent. Consumers save time because they don’t have to go to the store to make their purchases and companies save money by not owing physical stores and avoiding expensive rental payments. This method is convenient both for the company and for the consumer.

Nowadays most of the transactions take place online, meaning that credit cards and other online payment systems are involved. Image by mohamed Hassan from Pixabay Motivation

Credit card validator python project