A temporal consistency method for online review ranking

Providing appropriate online review ranking consistently with the entire review set is deemed important for e-commerce services to facilitate consumers decision making. Unlike the existing efforts that often treat online reviews statically, this paper takes the temporal dynamics of online reviews into account, and designs an effective method for time-aware review ranking. In doing so, first of all, a time-aware review consistency ranking (TRCR) problem is formulated, based on a newly defined metric, which aims to derive a compact review list with maximized expected consistency degree to the original review set. Furthermore, this problem is proven to be NP-hard, which leads us to developing an effective approximation by heuristically restricting the search space (i.e., TRCRea). This proposed method with related improvements show strengths on two aspects: one is that the informational decay of the reviews is well addressed at both macro and micro levels; and the other is that the compact review list provided to the consumers is obtained from a combined perspective of consistency and time-awareness in light of product features and sentiment orientations. Finally, the experiments on real-world data demonstrate the effectiveness and efficiency of the proposed method over baseline methods.