Formal Concept Analysis
Formal Concept Analysis (FCA) is a method to extract hierarchical "concepts" from the relationship between objects (subjects) and attributes (features) to obtain a structure called a "concept lattice.
- (1) Prepare the target set and feature (attribute) set.
- (2) Define "concepts" (groups of objects and their common features) based on the features that the objects possess.
- (3) Aligning these concepts by inclusion relations yields a hierarchical structure between concepts as "Concept Lattice". This concept lattice is useful for navigation and knowledge discovery between conceptual levels of abstraction by structuring data hierarchically based on feature sets.
(1) Organize the object (thing) and its attributes (characteristics): - For example, it is like preparing a table that lists the objects "apple," "banana," and "strawberry" and their attributes such as "red," "sweet," and "has skin. (2) Group them by commonalities: - Groups (concepts) are created by combining attributes common to the subject, such as "red and sweet fruits." (3) Arrange such groups hierarchically: - The groups with more attributes (more detailed and limited concepts) are placed lower, and the groups with fewer attributes (broader concepts) are placed higher, so that "which groups are included in which groups" are organized and visible.
relevance - biclustering
This page is auto-translated from [/nishio/Formal Concept Analysis](https://scrapbox.io/nishio/Formal Concept Analysis) using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.