IN-DEDUCTIVE and DAG-Tree Approaches for Large-Scale Extreme Multi-label Hierarchical Text Classification

Mohammad Golam Sohrab, Makoto Miwa, Yutaka Sasaki


This paper presents a large-scale extreme multi-label hierarchical text classification method that employs a large-scale hierarchical inductive learning and deductive classification (IN-DEDUCTIVE) approach using different efficient classifiers, and a DAG-Tree that refines the given hierarchy by eliminating nodes and edges to generate a new hierarchy. We evaluate our method on the standard hierarchical text classification datasets prepared for the PASCAL Challenge on Large-Scale Hierarchical Text Classification (LSHTC). We compare several classification algorithms on LSHTC including DCD-SVM, SVMperf, Pegasos, SGD-SVM, and Passive Aggressive, etc. Experimental results show that IN-DEDUCTIVE approach based systems with DCD-SVM, SGD-SVM, and Pegasos are promising and outperformed other learners as well as the top systems participated in the LSHTC3 challenge on Wikipedia medium dataset. Furthermore, DAG-Tree based hierarchy is effective especially for very large datasets since DAG-Tree exponentially reduce the amount of computation necessary for classification. Our system with IN-DEDUCIVE and DAG-Tree approaches outperformed the top systems participated in the LSHTC4 challenge on Wikipedia large dataset.


Hierarchical text classification; multi-label learning; indexing; extreme classification; tree-structured class hierarchy; DAG- or DG-structured class hierarchy

Full Text: PDF


  • There are currently no refbacks.