Frequent pattern growth（频繁模式增长（Frequent pattern growth））¶
Supported in: Batch
Frequent pattern (fp) growth finds frequent patterns in your dataset.
Transform categories: Aggregate, Other
Declared arguments¶
- Input dataset: Source dataset containing items columns and transaction column.
Table - Items column: Array column containing the items for the patterns.
Column\ - Minimum support: Minimum fraction of how often a pattern needs to be present.
Literal\
Examples¶
Example 1: Base case¶
Argument values:
- Input dataset: ri.foundry.main.dataset.a
- Items column:
customer_attributes - Minimum support: 0.6
Input:
| customer_attributes |
|---|
| [ age_group: 20-30, country: Germany, gender: Female ] |
| [ age_group: 20-30, country: Germany, gender: Male ] |
Output:
| pattern | pattern_occurrence | total_count |
|---|---|---|
| [ country: Germany, age_group: 20-30 ] | 2 | 2 |
| [ age_group: 20-30 ] | 2 | 2 |
| [ country: Germany ] | 2 | 2 |
Example 2: Null case¶
Argument values:
- Input dataset: ri.foundry.main.dataset.a
- Items column:
customer_attributes - Minimum support: 0.0
Input:
| customer_attributes |
|---|
| null |
Output:
| pattern | pattern_occurrence | total_count |
|---|---|---|
Example 3: Null case¶
Argument values:
- Input dataset: ri.foundry.main.dataset.a
- Items column:
customer_attributes - Minimum support: 0.0
Input:
| customer_attributes |
|---|
| [ age_group: 20-30, country: Germany, gender: Female ] |
| [ null ] |
Output:
| pattern | pattern_occurrence | total_count |
|---|---|---|
| [ country: Germany ] | 1 | 2 |
| [ country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ null ] | 1 | 2 |
| [ age_group: 20-30 ] | 1 | 2 |
| [ gender: Female ] | 1 | 2 |
| [ gender: Female, country: Germany ] | 1 | 2 |
| [ gender: Female, country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ gender: Female, age_group: 20-30 ] | 1 | 2 |
Example 4: Edge case¶
Argument values:
- Input dataset: ri.foundry.main.dataset.a
- Items column:
customer_attributes - Minimum support: 0.0
Input:
| customer_attributes |
|---|
| [ age_group: 20-30, country: Germany, gender: Female ] |
| [ age_group: 20-30, country: Germany, gender: Male ] |
Output:
| pattern | pattern_occurrence | total_count |
|---|---|---|
| [ gender: Male ] | 1 | 2 |
| [ gender: Male, country: Germany ] | 1 | 2 |
| [ gender: Male, country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ gender: Male, age_group: 20-30 ] | 1 | 2 |
| [ age_group: 20-30 ] | 2 | 2 |
| [ country: Germany ] | 2 | 2 |
| [ country: Germany, age_group: 20-30 ] | 2 | 2 |
| [ gender: Female ] | 1 | 2 |
| [ gender: Female, country: Germany ] | 1 | 2 |
| [ gender: Female, country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ gender: Female, age_group: 20-30 ] | 1 | 2 |
中文翻译¶
频繁模式增长（Frequent pattern growth）¶
支持：批处理（Batch）
频繁模式增长（Frequent pattern growth，FP-growth）用于发现数据集中的频繁模式。
转换类别：聚合（Aggregate）、其他（Other）
声明参数（Declared arguments）¶
- 输入数据集（Input dataset）： 包含项目列和事务列的源数据集。
表（Table） - 项目列（Items column）： 包含模式中项目的数组列。
列\<数组\<字符串>>（Column\ - 最小支持度（Minimum support）： 模式需要出现的最小比例。
字面量\<双精度浮点数>（Literal\）
示例¶
示例 1：基本情况¶
参数值：
- 输入数据集： ri.foundry.main.dataset.a
- 项目列：
customer_attributes - 最小支持度： 0.6
输入：
| customer_attributes |
|---|
| [ age_group: 20-30, country: Germany, gender: Female ] |
| [ age_group: 20-30, country: Germany, gender: Male ] |
输出：
| pattern | pattern_occurrence | total_count |
|---|---|---|
| [ country: Germany, age_group: 20-30 ] | 2 | 2 |
| [ age_group: 20-30 ] | 2 | 2 |
| [ country: Germany ] | 2 | 2 |
示例 2：空值情况¶
参数值：
- 输入数据集： ri.foundry.main.dataset.a
- 项目列：
customer_attributes - 最小支持度： 0.0
输入：
| customer_attributes |
|---|
| null |
输出：
| pattern | pattern_occurrence | total_count |
|---|---|---|
示例 3：空值情况¶
参数值：
- 输入数据集： ri.foundry.main.dataset.a
- 项目列：
customer_attributes - 最小支持度： 0.0
输入：
| customer_attributes |
|---|
| [ age_group: 20-30, country: Germany, gender: Female ] |
| [ null ] |
输出：
| pattern | pattern_occurrence | total_count |
|---|---|---|
| [ country: Germany ] | 1 | 2 |
| [ country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ null ] | 1 | 2 |
| [ age_group: 20-30 ] | 1 | 2 |
| [ gender: Female ] | 1 | 2 |
| [ gender: Female, country: Germany ] | 1 | 2 |
| [ gender: Female, country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ gender: Female, age_group: 20-30 ] | 1 | 2 |
示例 4：边界情况¶
参数值：
- 输入数据集： ri.foundry.main.dataset.a
- 项目列：
customer_attributes - 最小支持度： 0.0
输入：
| customer_attributes |
|---|
| [ age_group: 20-30, country: Germany, gender: Female ] |
| [ age_group: 20-30, country: Germany, gender: Male ] |
输出：
| pattern | pattern_occurrence | total_count |
|---|---|---|
| [ gender: Male ] | 1 | 2 |
| [ gender: Male, country: Germany ] | 1 | 2 |
| [ gender: Male, country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ gender: Male, age_group: 20-30 ] | 1 | 2 |
| [ age_group: 20-30 ] | 2 | 2 |
| [ country: Germany ] | 2 | 2 |
| [ country: Germany, age_group: 20-30 ] | 2 | 2 |
| [ gender: Female ] | 1 | 2 |
| [ gender: Female, country: Germany ] | 1 | 2 |
| [ gender: Female, country: Germany, age_group: 20-30 ] | 1 | 2 |
| [ gender: Female, age_group: 20-30 ] | 1 | 2 |