Review Graph Based Online Store Review Spammer Detection

Abstract

Abstract-Online reviews provide valuable information about products and services to consumers. However, spammers are joining the community trying to mislead readers by writing fake reviews. Previous attempts for spammer detection used reviewers' behaviors, text similarity, linguistics features and rating patterns. Those studies are able to identify certain types of spammers, e.g., those who post many similar reviews about one target entity. However, in reality, there are other kinds of spammers who can manipulate their behaviors to act just like genuine reviewers, and thus cannot be detected by the available techniques. In this paper, we propose a novel concept of a heterogeneous review graph to capture the relationships among reviewers, reviews and stores that the reviewers have reviewed. We explore how interactions between nodes in this graph can reveal the cause of spam and propose an iterative model to identify suspicious reviewers. This is the first time such intricate relationships have been identified for review spam detection. We also develop an effective computation method to quantify the trustiness of reviewers, the honesty of reviews, and the reliability of stores. Different from existing approaches, we don't use review text information. Our model is thus complementary to existing approaches and able to find more difficult and subtle spamming activities, which are agreed upon by human judges after they evaluate our results. 1

Fake Review Detection From a Product Review Using Modified Method of Iterative Computation Framework

“…The algorithm developed by Wang, et al, (2011) will converges very fast. Therefore, from 855928 review data, not all of it being processed with these two models.…”

“…While the aim of the performance test is to compare results from the ICF ++ method with another fake review detection methods. Method from Wang, et al, (2011) is reimplemented using java to see the performance. Then the fake review detection methods in Wang's research used as a reference comparison with method ICF ++.…”

“…But on the other hand, spammers are able to learn from their mistakes and change their tactics and eventually adapt to avoid antispam techniques that are already equipped with these indicators, so that in the end, the algorithm that utilized duplication approach no longer can be used to detect fake reviews anymore. Motivated by the difficulties in manually labeling training data, Wang et al,(2011) developed an unsupervised method. Computational methods were developed to measure the trustines of reviewers, the honesty of reviews, and the reliability of stores.…”

Spam Detection Framework Using Sentimenatl Analysis

“…Wang et al [7]. Likewise make a system of clients, audits and things and utilize essential suppositions (for instance an analyst is more dependable on the off chance that he/she composes more genuine reviews) and mark surveys.…”

Spotting Fake Reviews via Collective Positive-Unlabeled Learning

“…Inspired by the work from researchers who proposed graphbased models [1], [2], [9] and [12], we believe that exploiting the subtlety of the correlations between users, reviews and IP addresses would achieve better prediction results. Thus we first propose a collective classification algorithm MHCC (Multi-typed Heterogeneous Collective Classification) to identify fake reviews in our defined heterogeneous network over users, reviews and IP addresses.…”