Enhancement of Target-Oriented Opinion Words Extraction with Multiview-Trained Machine Reading Comprehension Model

<table class="table-group" id="tab5"><tr><td><table class="table"><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr class="thead"><td class="align_left">Views</td><td class="align_center">Question template</td></tr><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr><td class="align_left">TOWE</td><td class="align_center">1. Opinion words oriented to aspect target [TARGET]?</td></tr><tr><td class="align_left">OATE</td><td class="align_center">2. Aspect targets related to opinion word [OPINION]?</td></tr><tr><td class="align_left">PRC</td><td class="align_center">3. Is opinion word [OPINION] oriented to aspect target [TARGET]?</td></tr><tr class="table-tr"><td colspan="2"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Templates to generate questions. “[TARGET]” and “[OPINION]” are argument placeholders to be filled with aspect target and opinion phrases, respectively.</div>

Computational Intelligence and Neuroscience

tab5

Table 5

Table 5: Enhancement of Target-Oriented Opinion Words Extraction with Multiview-Trained Machine Reading Comprehension Model 