---
title: "FP-Growth advantages and disadvantages"
subject: "Data Mining"
chapter: 4
key: "semester_7/data_mining:fp-growth-pros-cons"
source: https://www.bctengineer.com/q/semester_7/data_mining/fp-growth-pros-cons
site: BCT Engineering
---

# FP-Growth advantages and disadvantages

Data Mining · Chapter 4: Association Analysis

- Asked in: year not given
- Video: [FP growth Vs Apriori Algorithm|FP growth tree vs Apriori algorithm in frequent pattern mining](https://www.youtube.com/watch?v=qaTC0Y_evbk) — Unfold Data Science
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_7/data_mining/fp-growth-pros-cons
- Whole chapter: https://www.bctengineer.com/md/semester_7/data_mining/chapter-4

## Question

Discuss advantages and disadvantages of FP-Growth algorithm.

## Short answer

Advantages: no candidate generation, two database scans, compact FP-tree, faster than Apriori, divide-and-conquer finds long patterns efficiently, scalable. Disadvantages: FP-tree may not fit memory for large/sparse data, complex to implement, not suitable for incremental data, requires rebuilding if min_sup changes.

## Answer

#### Advantages
- **No candidate generation**: avoids the huge candidate sets of Apriori.
- **Only two database scans**: one for counting items, one for building the tree.
- **Compact structure**: the FP-tree compresses the data by sharing common prefixes, so mining is done on a small tree in memory.
- **Much faster** than Apriori, especially for dense data and low minimum support (often by an order of magnitude).
- Divide-and-conquer approach finds **long frequent patterns** efficiently; it is complete and does not lose information about the frequent itemsets.
- Scales well; can be extended to parallel and database-projection versions.

#### Disadvantages
- The FP-tree may **not fit in main memory** for very large or very sparse databases (little prefix sharing, a tree almost as big as the data).
- **Complex** to implement and to understand compared with Apriori.
- Building and recursively mining conditional FP-trees costs time and memory, with many recursive structures.
- The tree is built for a **given minimum support**; if min_sup changes, it may need rebuilding.
- Not suitable for incremental or streaming data without modification (inserting a new transaction can change the item order).
- Gives only frequent itemsets; rule generation is a separate step.

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. Why does FP-growth avoid Apriori's huge candidate sets?
   A. Patterns grow from frequent items in the tree without generating candidates
   B. It lowers the minimum support
   C. It samples the database
   D. It counts only 1-itemsets
   Answer: A. Pattern growth works from conditional trees, not candidates. (Key point: Candidate generation)

2. In which situation is FP-growth especially faster than Apriori?
   A. Very small databases with high support
   B. Dense data and low minimum support
   C. Streaming data
   D. Data with no repeated items
   Answer: B. The tree compresses dense data well, and Apriori suffers at low support. (Key point: Speed)

3. What is a main disadvantage of FP-growth?
   A. It needs one scan per level
   B. It misses long patterns
   C. The FP-tree may not fit in memory for large or sparse data
   D. It cannot find complete results
   Answer: C. Sparse data shares few prefixes, so the tree is almost as big as the data. (Key point: Memory limitation)

4. What happens if min_sup changes?
   A. Nothing; the tree works for any support
   B. Only the header table changes
   C. The algorithm switches to Apriori
   D. The tree may need to be rebuilt
   Answer: D. The tree is built after dropping items below the given support. (Key point: Changing support)

## Related in Data Mining

- [FP-Growth and Apriori disadvantage](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Afp-growth-vs-apriori) — Data Mining, chapter 4
- [FP-tree on HotDogs data](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Afp-tree-hotdogs) — Data Mining, chapter 4
- [FP-tree on t1-t5 data](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Afp-tree-t1-t5) — Data Mining, chapter 4
- [ID3 and TP rate, FP rate, precision](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aid3-weka-tp-fp-precision) — Data Mining, chapter 3
- [ID3 steps, overfitting, TP/FP/recall](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aid3-steps-overfitting-recall) — Data Mining, chapter 3

## Related in other subjects

- [Importance of growth function; Theta, O, Omega](https://www.bctengineer.com/question.md?key=semester_4%2Fdata_structure_and_algorithm%3Agrowth-function-importance) — Data Structure and Algorithm, chapter 9