PeriodicCorrelatedPatternMining

class geoanalytics.patternMining.PeriodicCorrelatedPatternMining.PeriodicCorrelatedPatternMining(inputFile: str)[source]

Bases: PatternMiner

About this algorithm

Description:

This module implements the EPCPGrowth algorithm for mining periodic correlated patterns from temporal transactional databases. The algorithm discovers itemsets that appear frequently with strong correlation within specific periodic intervals controlled by periodic correlation thresholds.

Parameters:

  • inputFile (str): Path to the temporal transactional database file.

Attributes:

  • inputFile (str): The temporal transactional input file provided during object instantiation.

  • miner (EPCPGrowth): Instance of the EPCPGrowth algorithm from the PAMI library.

Execution methods

Calling from a Python program

from geoanalytics.patternMining import PeriodicCorrelatedPatternMining

miner = PeriodicCorrelatedPatternMining("data/input.txt")

miner.run(minSupport=3, minAllConf=0.6, maxPerAllConf=0.8, maxPer=10)

Credits

Written by M. Charan Teja, under the guidance of Professor Rage Uday Kiran.

run(minSupport: int, minAllConf: float, maxPerAllConf: float, maxPer: int | float)[source]

Executes the EPCPGrowth algorithm to mine periodic correlated patterns.

Parameters:

  • minSupport (int) – Minimum support threshold for frequent itemsets.

  • minAllConf (float) – Minimum all-confidence threshold for correlation.

  • maxPerAllConf (float) – Maximum periodic all-confidence threshold controlling periodic correlation.

  • maxPer (int or float) – Maximum periodicity controlling pattern recurrence interval.

Output:

Prints the discovered periodic correlated patterns to the console.