In this tutorial, we will explore the Credit Card Fraud Detection dataset available publicly on Kaggle. It is very crucial for credit card companies to detect fraud to avoid making losses and at the same time, not charging the customers for what they did not actually themselves purchase. We will use GridDB to extract the data, followed by building machine learning models to accurately detect frauds.<\/p>\n
The outline of the tutorial is as follows:<\/p>\n
While loading the dataset, this tutorial will cover two methods — Using GridDB as well as Using Pandas. To access GridDB using Python, the following packages also need to be installed beforehand:<\/p>\n
The dataset contains all the credit card transactions made by European cardholders on two days in the month of September 2013. There are 284,407 transactions in total out of which 492 are frauds. As evident, the dataset is highly imbalanced, as the fraud class comprises only of 0.172% of all the transactions.<\/p>\n
A PCA (Principal Component Analysis) transformation has been applied on the dataset. PCA transformation refers to conversion of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. PCA gives V1, V2, …, V28 as the features and the only features that have not been transformed are Time of the transaction, and Amount of the transaction. Since the rest of the raw (original) features contain confidential information, these are not provided in the final dataset.<\/p>\n
Description of the features:<\/p>\n
V1, V2, …, V28: Transformed features, hence they might not necessarily be raw features related to credit card transactions.<\/p>\n<\/li>\n<\/ul>\n
The dataset is available publicly and can be downloaded from Kaggle<\/a>. Please go ahead and download the dataset to follow along this tutorial.<\/p>\n In case you run into a package installation error, you can install it by typing Let’s proceed and load the dataset into our notebook.<\/p>\n To store large amounts of data, a CSV file can be cumbersome. GridDB serves as a perfect alternative as it in open-source and a highly scalable database. GridDB is a scalable, in-memory, No SQL database which makes it easier for you to store large amounts of data. If you are new to GridDB, a tutorial on reading and writing to GridDB<\/a> can be useful.<\/p>\n Assuming that you have already set up your database, we will now write the SQL query in python to load our dataset.<\/p>\n Note that the We can also use Pandas’ Once the dataset is loaded, let us now explore the dataset. We’ll print the first 10 rows of this dataset using head() function passing in 10 as argument, indicating the number of rows we want to print from the top.<\/p>\n2. Importing Required Libraries<\/h2>\n
import numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\n\nfrom sklearn.neighbors import KNeighborsClassifier\nfrom sklearn.model_selection import train_test_split\n\nfrom sklearn.metrics import precision_recall_curve\nfrom sklearn.metrics import auc\nfrom sklearn.metrics import make_scorer\nfrom sklearn.metrics import classification_report\n\nfrom sklearn.model_selection import cross_val_score<\/code><\/pre>\n<\/div>\npip install package-name<\/code> in the command line. Alternatively, if you’re using a conda virtual environment, you can type conda install package-name<\/code>.<\/p>\n3. Loading the Dataset<\/h2>\n
3.a Using GridDB<\/h4>\n
import griddb_python as griddb\n\nsql_statement = ('SELECT * FROM credit_card_dataset')\nheart_dataset = pd.read_sql_query(sql_statement, cont)<\/code><\/pre>\n<\/div>\ncont<\/code> variable has the container information where our data is stored. Replace the credit_card_dataset<\/code> with the name of your container. More info can be found in this tutorial reading and writing to GridDB<\/a>.<\/p>\n3.b Using Pandas<\/h4>\n
read_csv<\/code> function to load our data. Both of the above methods will lead to the same output as the data is loaded in the form of a pandas dataframe using either of the methods.<\/p>\ncredit_card_dataset = pd.read_csv('creditcard.csv')<\/code><\/pre>\n<\/div>\n4. Exploratory Data Analysis<\/h2>\n
credit_card_dataset.head(10)<\/code><\/pre>\n<\/div>\n