Skip to content
HPI-Information-SystemsPublic

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

3 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FLAPS — Fast Lattice Accurate Pattern Search

FLAPS (Fast Lattice Accurate Pattern Search) is a Java implementation for efficiently identifying Maximal Uncovered Patterns (MUPs) in categorical datasets. MUPs capture regions of the data space that are underrepresented or entirely missing, making them essential for assessing data coverage and diversity in structured datasets.


Overview

FLAPS follows a two-stage approach:

  1. Seeder — identifies attribute–value combinations that are sufficiently covered and spreads this information through the lattice to prune the search space early.
  2. Validator — traverses the remaining unexplored regions to efficiently identify uncovered and maximal patterns.

Both stages execute sequentially.


Parameters

Parameter Description Example Default
--csv Path to the input dataset (CSV format). --csv data/airbnb.csv required
--uninteresting-attributes Comma-separated list of attributes to exclude from analysis (e.g., IDs). --uninteresting-attributes id,flight_number (none)
--interesting-attributes Comma-separated list of attributes to explicitly include. If omitted, all remaining attributes are used. --interesting-attributes airline,source_city,class (none)
--mincov Minimum coverage threshold (mincov) — defines when a pattern is considered covered. --mincov 8 1
--chunk-size Number of attributes per chunk used in the Seeder stage. --chunk-size 5 1
--chunks-per-row Number of chunks sampled per tuple during seeding. --chunks-per-row 10 1
--output-mups [path] Write reverse-mapped MUPs and exact coverage. Without a path, writes mups_DATASET_NAME_mincov_x.csv under results/. --output-mups (disabled)

Usage

Requires Java 21 or newer and Maven. Run from the project directory.

mvn clean package

java -jar target/diversity-assessment-1.0-SNAPSHOT.jar \
  --csv /path/to/dataset.csv \
  --mincov 1 \
  --chunk-size 14 \
  --chunks-per-row 15

Runtime summaries are written under results/.

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages