Mining Complex Hydrobiological Data with Galois Lattices

TL;DR

This paper demonstrates how Galois lattices can be applied to complex hydrobiological data by transforming it into binary form, enabling clustering of water plant species based on traits and modalities.

Contribution

The paper introduces methods for converting complex hydrobiological data into binary format and applies Galois lattices for clustering and relation discovery among traits.

Findings

Effective data transformation techniques for Galois lattices

Successful clustering of hydrobiological data

Insights into trait relationships among water plants

Abstract

We have used Galois lattices for mining hydrobiological data. These data are about macrophytes, that are macroscopic plants living in water bodies. These plants are characterized by several biological traits, that own several modalities. Our aim is to cluster the plants according to their common traits and modalities and to find out the relations between traits. Galois lattices are efficient methods for such an aim, but apply on binary data. In this article, we detail a few approaches we used to transform complex hydrobiological data into binary data and compare the first results obtained thanks to Galois lattices.