Please help me to understand with a draft ? There are eight different data approaches. These approaches are:

1. Classification
2. Regression
3. Similarity Matching
4. Clustering
5. Co-occurrence grouping
6. Profiling
7. Link Prediction, and
8. Data Reduction.

Alibaba uses the profiling data approach to identify potential cases of fraud. Alibaba has worked to capture fraud signals directly from its extensive database of user behaviors and its network, and then analyzes them in real-time using machine learning to accurately sort the bad users from the good ones.

Facebook uses link prediction to predict a relationship between two people when it suggests people that you may likely know based on similar other friends, high schools, college or work locations, etc.