The annotated corpus was processed with natural language processing tools and rule-based approaches were applied to the sentences for extracting the relations between TCM effecters and effects.
To extract the relations between effecters and effects from literature, sentences were part-of-speech (POS -tagged by Genia Tagger [ 31] and noun-POS -tagged chunks were identified by NPchunker [ 32].
These include optimizing dose control, limiting off-target effects, improving delivery of RNAi effecters and demonstration of efficacy in suitable animal models before embarking on clinical evaluations.
Rule 2.1 extracts effecter and effect NP chunks connected by active verbs.
The effecter and effect relations were extracted by matching preprocessed sentences with manually designed four rule patterns for high precision.
Rule 3.1 extracts effecter and effect NP chunks connected by past participle verbs that are frequently observed in English for describing research discoveries.
For instance, effecter, ' andP', and effect, ' antiviral', are able to be extracted from the sentence, ' APBP was shown to have potent antiviral activity'.
For example, the sentence, ' LZ-8, a new and recently discovered immunomodulator from Ganoderma lucidum, has been shown to have immunosuppressive activity' can be matched and effecter, ' LZ-8', and effect, ' immunosuppressive', are able to be extracted, respectively.
To evaluate the precision of the extracted relations between effecter and effect, 20 TCM entries were randomly selected in the test set and the precision of the 1185 relations is 91.2%.
Genetically modified crops and overuse of Roundup?
