Research of Deceptive Review Detection Based on Target Product Identification and Metapath Feature Weight Calculation

<table class="fixed-width table-group" id="tab2"><tr><td><table class="table"><colgroup><col style="width:8.74em"/><col style="width:5.74em"/><col style="width:4.62em"/></colgroup><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr class="thead"><td class="align_left">Features</td><td class="align_center">Reviewers</td><td class="align_center">Reviews</td></tr><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr><td class="align_left">Based on behavior</td><td class="align_center">MNRD  <br/>RPR  <br/>RNR  <br/>BST  <br/>ERD  <br/>BDS  <br/>RD  <br/>RWR</td><td class="align_center">RRD  <br/>ETF</td></tr><tr><td class="align_left" colspan="3"><hr/></td></tr><tr><td class="align_left">Based on semantic</td><td class="align_center">ACS</td><td class="align_center">RPP  <br/>ROW</td></tr><tr class="table-tr"><td colspan="3"><hr class="tbody-hr"/></td></tr></table></td></tr><tr class="table-fn"><td><div>MNRD: max number of reviews daily, RPR: ratio of positive reviews, RNR: ratio of negative reviews, BST: burstiness, ERD: entropy of ratings distribution, BDS: brand deviation score, RD: rating deviation, RWR: ratio of weekend reviews, RRD: review rating deviation, ETF: early time frame, ACS: average content similarity, RPP: ratio of 1st and 2nd person pronouns, and ROW: ratio of objective words.<br/></div></td></tr></table>

<div>Features extraction in different strategy.</div>

Complexity

tab2

Table 2

Table 2: Research of Deceptive Review Detection Based on Target Product Identification and Metapath Feature Weight Calculation 