Evolutionary hypernetwork classifiers for protein-proteininteraction sentence filtering

Abstract

Protein-Protein Interaction (PPI) extraction, among ongoing biomedical text mining challenges, is becoming a topic in focus because of its crucial role in providing a starting point to understand biological processes. Machine learning (ML) techniques have been applied to extract the PPI information from biomedical literature. Although they have provided reasonable performance so far, more features are required for real use. In particular, many ML-approaches lack human understandability for learned models. Here, we propose a novel method for classifying PPI sentences. Our approach utilizes the modified hypernetwork model, a hypergraph with weighted hyperedges that are calibrated via an evolutionary learning method. The evolutionary hypernetwork memorizes fragments of training patterns while selfadjusting its own structure for detecting PPI sentences. For experiments, we show that our approach provides competitive performance compared to other ML methods. Apart from its superior classification performance, the evolving hypernetwork model comes with a highly interpretable structure. We show how significant PPI patterns can be naturally extracted from the learned model. We also analyze the discovered patterns.

Evolutionary Learning of Syntax Patterns for Genic Interaction Extraction

“…A very good and broad review of evolutionary approaches to NLP is [2]. The binary classification of sentences that we consider here, though, has received very little attention so far and we are aware of only one evolutionary proposal in this area [7]. The cited work advocates the use of a probabilistic model called the hypernetwork classifier meant to capture the correlation between sets of words and the class in which a sentence containing those words belongs to.…”

A molecular evolutionary algorithm for learning hypernetworks on simulated DNA computers

“…Each hyperedge contains several vertices as features. The hypernetworks is trained by an evolutionary process using the DNA molecular operations of matching, selection, and amplification [15][16] [17]. The hyperedges in hypernetworks are the molecular individuals.…”

Non-uniform Evolving Hypergraphs and Weighted Evolving Hypergraphs

“…They combined intuitionistic fuzzy theory with the hypergraph concept and defined several intuitionistic fuzzy structures, which are more flexible than the classic models. Bootkrajang et al 20. built a model of associative memory based on an undirected hypergraph of weighted edges.…”