Formal context reduction in deriving concept hierarchies from corpora using adaptive evolutionary clustering algorithm star

Abstract It is beneficial to automate the process of deriving concept hierarchies from corpora since a manual construction typically time-consuming and resource-intensive process. As such, overall learning encompasses set steps: parsing text into sentences, splitting sentences then tokenising it. After lemmatisation step, pairs are extracted using formal context analysis (FCA). However, there might be some uninteresting erroneous in context. Generating may lead process, so size reduction require remove uninterested pairs, taking less time extract lattice accordingly. In this premise, study aims propose two frameworks: (1) A framework review current corpus utilising (FCA); (2) decrease context’s ambiguity first an adaptive version evolutionary clustering algorithm (ECA*). Experiments conducted by applying 385 sample Wikipedia on frameworks examine reducing context, which leads yield hierarchy. The resulting evaluated standard one lattice-invariants. Accordingly, homomorphic between lattices preserves quality 89% contrast basic ones, reduced inherits structural relation one. ECA* examined against its four counterpart baseline algorithms (Fuzzy K-means, JBOS approach, AddIntent algorithm, FastAddExtent) measure execution random datasets with different densities (fill ratios). results show that performs faster than other mentioned competitive techniques fill ratios.