Multistage non-deterministic classification using secondary concept graphs and graph convolutional networks for high-level feature extraction

TL;DR

This paper introduces a multi-stage non-deterministic classification approach using secondary concept graphs and GCNs to improve high-level feature extraction and classification accuracy in complex graph-structured data.

Contribution

It proposes a novel multi-stage method combining GCNs and conceptual graphs for enhanced high-level feature extraction and more accurate classification in graph data.

Findings

Achieved 96% accuracy on Cora dataset

Achieved 93% accuracy on Citeseer dataset

Achieved 95% accuracy on PubMed dataset

Abstract

Graphs, comprising nodes and edges, visually depict relationships and structures, posing challenges in extracting high-level features due to their intricate connections. Multiple connections introduce complexities in discovering patterns, where node weights may affect some features more than others. In domains with diverse topics, graph representations illustrate interrelations among features. Pattern discovery within graphs is recognized as NP-hard. Graph Convolutional Networks (GCNs) are a prominent deep learning approach for acquiring meaningful representations by leveraging node connectivity and characteristics. Despite achievements, predicting and assigning 9 deterministic classes often involves errors. To address this challenge, we present a multi-stage non-deterministic classification method based on a secondary conceptual graph and graph convolutional networks, which includes distinct steps: 1) leveraging GCN for the extraction and generation of 12 high-level features: 2) employing incomplete, non-deterministic models for feature extraction, conducted before reaching a definitive prediction: and 3) formulating definitive forecasts grounded in conceptual (logical) graphs. The empirical findings indicate that our proposed approach outperforms contemporary methods in classification tasks. Across three datasets Cora, Citeseer, and PubMed the achieved accuracies are 96%, 93%, and 95%, respectively. Code is available at https://github.com/MasoudKargar.