Introduction

A Gradual Pattern (GP) is a co-occurring set of gradual items (GI) that captures covariations between attributes. A pattern’s quality is measured quantitatively by its computed support value.

Illustrative Example

Consider a dataset containing 6 objects across 3 features (Age, Salary, and Cars):

Object

Age

Salary

Cars

o1

23

52,000

0

o2

27

51,000

1

o3

31

50,000

1

o4

36

48,000

1

o5

40

47,000

2

o6

40

45,000

2

An extracted GP might take the following form:

{ Age+, Salary- }      [Support ≈ 0.83]

This expression reveals that in 83.3% of the dataset (5 out of 6 objects), a strict increase in Age (+) strongly correlates with a simultaneous decrease in Salary (-).

Step-by-Step Validation:

  • Comparing o1 → o2: Age increases (23 → 27), Salary decreases (52k → 51k). (Valid)

  • Comparing o2 → o3: Age increases (27 → 31), Salary decreases (51k → 50k). (Valid)

  • Comparing o3 → o4: Age increases (31 → 36), Salary decreases (50k → 48k). (Valid)

  • Comparing o4 → o5: Age increases (36 → 40), Salary decreases (48k → 47k). (Valid)

  • Comparing o5 → o6: Age stays the same (40 → 40), but Salary decreases (47k → 45k). Depending on your variation definition (strict vs. non-strict inequality), this step sequence validates 5 out of 6 objects.

Installation

The library is available on PyPI. To install it, run the following command in your terminal:

pip install so4gp

Basic Usage

After installing the so4gp package, you can import it as follows:

import so4gp as sgp

The sgp namespace contains all necessary classes, functions, and algorithms. Classes and functions are accessible via sgp.ClassName or sgp.function_name.

To use the algorithms, import them via:

from so4gp.algorithms import GRAANK, AntGRAANK, GeneticGRAANK, ClusterGP, TGradAMI

The so4gp algorithms require a numeric dataset provided as either a pandas.DataFrame or a path to a CSV file.

All so4gp functions and classes are documented in the API Section.

References

  • Owuor, D., Runkler T., Laurent A., Menya E., Orero J (2021), Ant Colony Optimization for Mining Gradual Patterns. International Journal of Machine Learning and Cybernetics. https://doi.org/10.1007/s13042-021-01390-w

  • Dickson Owuor, Anne Laurent, and Joseph Orero (2019). Mining Fuzzy-temporal Gradual Patterns. In the proceedings of the 2019 IEEE International Conference on Fuzzy Systems (FuzzIEEE). IEEE. https://doi.org/10.1109/FUZZ-IEEE.2019.8858883

  • Laurent A., Lesot MJ., Rifqi M. (2009) GRAANK: Exploiting Rank Correlations for Extracting Gradual Itemsets. In: Andreasen T., Yager R.R., Bulskov H., Christiansen H., Larsen H.L. (eds) Flexible Query Answering Systems. FQAS 2009. Lecture Notes in Computer Science, vol 5822. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-04957-6_33