|
| Authors | Technique | Domain area | Type of relations | Year of publication | Reported results |
|
|
Huang et al. [20] | Hybrid approach (shallow parsing and pattern matching) | Biomedical | Protein-protein (P2P) interaction | 2006 | 80% -score |
|
| Frunza and Inkpen [21] | Hybrid approach | Biomedical | Disease and treatment relation (cure, prevent, and side effect relations) | 2010 | Accuracy Cure 95%
Prevent relation 75%
Side effect 46% |
|
| Sharma et al. [22] | Verb-centric algorithm | Biomedical | Not mentioned | 2010 | 90% -score |
|
| Ben Abacha and Zweigenbaum [23] | Hybrid approach (pattern based and machine learning) | Biomedical | Disease and treatment relation | 2011 | 94.07% -score |
|
| Ben Abacha and Zweigenbaum [24] | Linguistic patterns and domain knowledge | Biomedical | Relation between 16 entities | 2011 | Precision of 75.72% and recall of 60.64% |
|
| Yang et al. [25] | Verb-centric approach | Biomedical | Relation between (foods, chemicals, diseases, proteins, and genes) | 2011 | 90.5% -score |
|
| Kadir and Bokharaeian [26] | Hybrid approach (rule-based, kernel based, and cooccurrence based methods) | Biomedical | Not mentioned | 2013 | Not reported |
|
| Rosario and Hearst [19] | Graphical models and neural network | Biomedical | Disease and treatment relation (cure, prevent, side effect relations) | 2004 | Accuracy Cure 92.6%
Prevent relation 38.5%
Side effect 20% |
|
