Table 1.
Performance Summary of Ontology Learning Techniques
| Techniques | Domain | Performance | References | ||
|---|---|---|---|---|---|
| Paper | Tools | ||||
| Linguistic Techniques | |||||
| Preprocessing | Berkley Parser | Tourism, Sport | Precision=95.7% | (28) | Text2Onto(75, 120, 121, 122) (http://neon-toolkit.org/wiki/1.x/Text2Onto.html), CRCTOL (28), https://nlp.stanford.edu/software/lex-parser.shtml, http://nlp.cs.berkeley.edu/ |
| Stanford Parser | Precision=90.3% | ||||
| Syntactic Analysis for headword modifier | Chinese Text | Accuracy=83.3% | (29) | https://github.com/kimduho/nlp/wiki/Head-modifier-principle-(or-relation) | |
| Relation Extraction | Lexico-syntactic Parsing | News | Accuracy=75.5% | (40) | Text2Onto (75, 120, 121, 122) (http://neon-toolkit.org/wiki/1.x/Text2Onto.html), CRCTOL (28), ASIUM (117, 118, 119) (http://www-ai.ijs.si/∼ilpnet2/systems/asium.html), TextStorm/Clouds (27, 123) |
| Dependency Analysis | Bioinformatics | Accuracy=83.3% | (38) | ||
| Statistical Techniques | |||||
| Term Extraction | C/NC Value | Medical | Precision=89.7% | (26) | OntoGain (72), https://github.com/Neuw84/CValue-TermExtraction |
| Computer Science | Precision=86.67% | ||||
| Contrastive Analysis | Chinese Text | Precision=70% | (56) | OntoLearn (49, 124, 55, 125), CRCTOL (28), OntoGain (72) | |
| Co-occurrence Analysis | Biomedical (Cancer) | Precision=67.3% | (62) | Text2Onto (75, 120, 121, 122) (http://neon-toolkit.org/wiki/1.x/Text2Onto.html), https://github.com/gsi-upm/sematch | |
| Clustering | Tourism | Accuracy=68.52% | (66) | ASIUM (117, 118, 119) (http://www-ai.ijs.si/∼ilpnet2/systems/asium.html), Text2Onto (75, 120, 121, 122) (http://neon-toolkit.org/wiki/1.x/Text2Onto.html), https://pythonprogramminglanguage.com/kmeans-text-clustering/ | |
| Tourism | Accuracy=53.2% | ||||
| Relation Extraction | Formal Concept Analysis | Medical | Precision=47% | (72) | OntoGain (72), https://github.com/xflr6/concepts |
| Computer Science | Precision=44% | ||||
| Hierarchical Clustering | Medical | Precision=71% | (72) | Text2Onto (75, 120, 121, 122) (http://neon-toolkit.org/wiki/1.x/Text2Onto.html), https://github.com/mstrosaker/hclust | |
| Cooking | Precision=92.1% | (71) | |||
| Finance | F1 Score=18.51% | (75) | |||
| Tourism | F1 Score=21.4% | (75) | |||
| Association Rule Mining | Medical | Accuracy=72.5% | (72) | Text2Onto (75, 120, 121, 122) (http://neon-toolkit.org/wiki/1.x/Text2Onto.html) | |
| Logical | |||||
| Inductive Logical Programming | English | Accuracy=96% | (83) | TextStorm/Clouds (27, 123) , Syndikate (126, 11), http://pyke.sourceforge.net/ | |