In this chapter, you'll learn about the typical challenges associated with fraud detection, and will learn how to resample your data in a smart way, to tackle problems with imbalanced data.

Introduction to fraud detection

Checking the fraud to non-fraud ratio

Plotting your data

Increasing successful detections using data resampling

Resampling methods for imbalanced data

Applying SMOTE

Compare SMOTE to original data

Fraud detection algorithms in action

Exploring the traditional way to catch fraud

Using ML classification to catch fraud

Logistic regression combined with SMOTE

Using a pipeline

Introduction and preparing your data

Now that you're familiar with the main challenges of fraud detection, you're about to learn how to flag fraudulent transactions with supervised learning. You will use classifiers, adjust them, and compare them to find the most efficient fraud detection model.

Review of classification methods

Natural hit rate

Random Forest Classifier - part 1

Random Forest Classifier - part 2

Performance evaluation

Performance metrics for the RF model

Plotting the Precision Recall Curve

Adjusting your algorithm weights

Model adjustments

Adjusting your Random Forest to fraud detection

GridSearchCV to find optimal parameters

Model results using GridSearchCV

Ensemble methods

Logistic Regression

Voting Classifier

Adjust weights within the Voting Classifier

Fraud detection using labeled data

This chapter focuses on using unsupervised learning techniques to detect fraud. You will segment customers, use K-means clustering and other clustering algorithms to find suspicious occurrences in your data.

Perilaku normal versus abnormal

Menjelajahi data Anda

Segmentasi pelanggan

Menggunakan statistik untuk mendefinisikan perilaku normal

Metode clustering untuk mendeteksi kecurangan

Menskalakan data

Pengelompokan K-means

Metode Elbow

Menetapkan penipuan versus bukan penipuan

Mendeteksi outlier

Memeriksa hasil model

Metode pendeteksian kecurangan lainnya dengan clustering

DBSCAN

Menilai klaster terkecil

Memeriksa hasil

Fraud detection using unlabeled data

In this final chapter, you will use text data, text mining, and topic modeling to detect fraudulent behavior.

Using text data

Word search with dataframes

Using list of terms

Creating a flag

Text mining to detect fraud

Removing stopwords

Cleaning text data

Topic modeling on fraud

Create dictionary and corpus

LDA model

Flagging fraud based on topics

Interpreting the topic model

Finding fraudsters based on topic

Recap

Fraud detection using text

Chapter 1 datasets

Chapter 2 datasets

Chapter 3 datasets

Chapter 4 datasets

Sebuah organisasi biasanya kehilangan sekitar 5% dari pendapatan tahunannya akibat kecurangan. Dalam kursus ini, Anda akan mempelajari cara memerangi kecurangan dengan menggunakan data. Misalnya, Anda akan belajar menerapkan algoritma supervised learning untuk mendeteksi perilaku curang yang mirip dengan kasus sebelumnya, serta metode unsupervised learning untuk menemukan jenis aktivitas kecurangan baru. Selain itu, dalam analitik kecurangan Anda sering berhadapan dengan himpunan data yang sangat tidak seimbang saat mengklasifikasikan antara kecurangan dan bukan kecurangan, dan sepanjang kursus ini Anda akan mempelajari beberapa teknik untuk menanganinya. Kursus ini memadukan wawasan teknis dan teoretis serta menunjukkan secara langsung cara mengimplementasikan model deteksi kecurangan secara praktis. Selain itu, Anda akan mendapatkan kiat dan saran dari pengalaman nyata untuk membantu Anda mencegah kesalahan umum dalam analitik kecurangan.

Unsupervised Learning in Python

Supervised Learning with scikit-learn

Pelajari deteksi penipuan dengan Python: resampling data, pembelajaran terawasi, segmentasi, K-means.

Perilaku normal versus abnormal

Create Your Free Account