The FP-Growth (Frequent Pattern Growth) algorithm is a popular method for mining frequent itemsets in transaction data, used in association rule learning (like market basket analysis). Unlike the classic Apriori algorithm, FP-Growth doesn’t generate candidate itemsets explicitly. Instead, it compresses the dataset into a compact structure called an FP-tree (Frequent Pattern Tree), which is a prefix tree of all transactions with counts. Then it extracts frequent itemsets by recursively growing patterns from this tree, following item frequency descending order. FP-Growth tends to be faster than Apriori for large datasets because it avoids the expensive candidate generation step and reduces database scans. After getting frequent itemsets, one can generate association rules (with confidence, lift, etc.). It’s an efficient approach to find all combinations of items that appear together frequently in a transactional dataset (e.g., “bread” and “butter” often bought together).