Dissecting AI-Generated Fake Reviews: Detection and Analysis of GPT-Based Restaurant Reviews on Social Media

2023 | International Conference on Information Systems | Citations: 0

Authors: Gambetti, Alessandro; Han, Qiwei

Abstract: Recent advances in generative models such as GPT may be used to fabricate indis ... Expand

Abstract: Recent advances in generative models such as GPT may be used to fabricate indistinguishable fake customer reviews at a much lower cost, posing challenges for social media platforms to detect this kind of content. This study addresses two research questions: (1) the effective detection of AI-generated restaurant reviews generated from high-quality elite authentic reviews, and (2) the comparison of out-of-sample predicted AI-generated reviews and authentic reviews across multiple dimensions of review, user, restaurant, and content characteristics. We fine-tuned a GPT text detector to predict fake reviews, significantly outperforming existing solutions. We applied the model to predict non-elite reviews that already passed the Yelp filtering system, revealing that AI-generated reviews typically score higher ratings, users posting such content have less established Yelp reputations and AI-generated reviews are more comprehensible and less linguistically complex than human-generated reviews. Notably, machine-generated reviews are more prevalent in low-traffic restaurants in terms of customer visits. Collapse

Topics: fake review artificial intelligence application programming interface open source pandemic

Methods: machine learning survey computational algorithm sensitivity analysis ANOVA

NOTION OF EXPLAINABLE ARTIFICIAL INTELLIGENCE - AN EMPIRICAL INVESTIGATION FROM A USER'S PERSPECTIVE

2023 | European Conference On Information Systems | Citations: 0

Authors: Haque, Bahalul; Islam, Najmul; Mikalef, Patrick

Abstract: ... Expand

Abstract: Collapse

Topics: recommender system explainable artificial intelligence user requirement artificial intelligence website

Methods: qualitative interview qualitative content analysis focus group group interview computational algorithm

Beyond a bunch of reviews: The quality and quantity of electronic word-of-mouth

2023 | Information & Management | Citations: 0

Authors: Hung, Shiu-Wan; Chang, Che-Wei; Chen, Ssu-Yu

Abstract: This work proposed an integrated model incorporating review quality and quantity ... Expand

Abstract: This work proposed an integrated model incorporating review quality and quantity to explore the extent to which consumers adopt electronic word-of-mouth (e-WOM) information. A linear structural equation model was employed to analyze 435 valid responses and test it against the hypotheses. The empirical results suggest that in terms of e-WOM the influence of review quantity outweighs that of the review quality. Review consistency negatively correlates with informational influence. Consumers are alert to highly consistent reviews and may easily doubt ones that contain similar content. Furthermore, the relationship between perceived persuasiveness and social influence will not be inhibited by confirmation bias. Collapse

Topics: online word of mouth social influence electronic commerce decision making social network

Methods: survey structural equation modeling statistical hypothesis test theory development data transformation

Theories: technology acceptance theory

To Brush or Not to Brush: Product Rankings, Consumer Search, and Fake Orders

2023 | Information Systems Research | Citations: 0

Authors: Jin, Chen; Yang, Luyi; Hosanagar, Kartik

Abstract: Brushing—online merchants placing fake orders of their own products—has been a w ... Expand

Abstract: Brushing—online merchants placing fake orders of their own products—has been a widespread phenomenon on major e-commerce platforms. One key reason why merchants brush is that it boosts their rankings in search results. Products with higher sales volume are more likely to rank higher. Additionally, rankings matter because consumers face search frictions and narrow their attention to only the few products that show up at the top. Thus, fake orders can affect consumer choice. We focus on this search-ranking aspect of brushing and build a stylized model to understand merchants’ strategic brushing behavior as well as how it affects consumers. We consider a high-type merchant (who sells a more popular product) and a low-type merchant (who sells a less popular product) competing on an e-commerce platform where product rankings evolve over time. We find that if brushing gets more costly for merchants (e.g., due to stricter platform policies), it may sometimes surprisingly harm consumers as it may only blunt brushing by the high-type merchant but intensify brushing by the low-type merchant. If search is less costly for consumers (e.g., due to improved search technologies), it may not always benefit consumers, either. Moreover, the design of the ranking algorithm is critical: placing more weight on sales-volume-related factors may trigger a nonmonotone change in consumer welfare; tracking recent sales only as opposed to cumulative sales does not always dial down brushing and, in fact, may sometimes cause the low-type merchant to brush more.History: Ravi Bapna, Senior Editor; De Liu, Associate Editor.Funding: C. Jin gratefully acknowledges the Singapore Ministry of Education Academic Research Fund [Tier 1, Grant R-253-000-144-133], The Wharton School Dean’s Postdoctoral Research Fund, and Mack Institute Research Fund. All authors gratefully acknowledge the 2019 Networks, Electronic Commerce and Telecommunications Institute (NET) Summer Research Grant.Supplemental Material: The online appendices are available at https://doi.org/10.1287/isre.2022.1128. Collapse

Topics: IS economics consumer surplus electronic commerce fake review electronic market

Methods: computational algorithm literature sample

Theories: economic search theory

The 2021 German Federal Election on Social Media: Analysing Electoral Risks Created by Twitter and Facebook

2023 | HICSS | Citations: 0

Authors: Kübler, Johanne; Sekwenz, Marie-Therese; Rachinger, Felicitas; König, Anna; Gsenger, Rita; Pírková, Eliška; Wagner, Ben; Kettemann, Matthias C.; Krennerich, Michael; Ferro, Carolina

Abstract: Safeguarding democratic elections is hard. Social media plays a vital role in th ... Expand

Abstract: Safeguarding democratic elections is hard. Social media plays a vital role in the discourse around elections and during electoral campaigns. The following article provides an analysis of the ‘systemic electoral risks’ created by Twitter and Facebook and the mitigation strategies employed by the platforms. It is based on the 2020 proposal by the European Commission for the new Digital Services Act (DSA) in the context of the 2021 German federal elections. This article focuses on Twitter and Facebook and their roles during the German federal elections that took place on 26 September 2021. We analysed three systemic electoral risk categories: 1) the dissemination of illegal content, 2) negative effects on electoral rights, and 3) the influence of disinformation and developed systematic categories for this purpose. In conclusion, we discuss how to respond to these challenges as well as avenues for future research. Collapse

Topics: Twitter Facebook disinformation social media application programming interface

Methods: literature study

Fake review detection system for online E-commerce platforms: A supervised general mixed probability approach

2023 | Decision Support Systems | Citations: 0

Authors: Luo, Jiwei; Luo, Jian; Nan, Guofang; Li, Dahui

Abstract: Online consumer reviews play an important role in helping consumers judge the qu ... Expand

Abstract: Online consumer reviews play an important role in helping consumers judge the quality and authenticity of products on e-commerce platforms. However, the constant presence of fake reviews on these platforms has significantly impacted the operation and development of e-commerce platforms. In this study, we develop a novel supervised probabilistic method to detect fake reviews by utilizing the difference in the distribution of nonfraudulent reviews and that of fake reviews. Specifically, we first derive the univariate distributions of several unique features (linguistic, behavioral, and interrelationship features). We then integrate these distributions into two mixed distributions according to their labels to represent the overall difference between non-fraudulent reviews and fake reviews. Next, we randomly generate synthetic review data points with different labels from the above mixed distributions. Finally, we train a Multilayer Perceptron model by using these synthetic review data to obtain a classifier. We conducted several experiments to test the model using several original real-world review datasets. Numerical results indicated that the proposed supervised method outperformed some wellknown sampling models and fake review detection methods, in terms of classification accuracy. Moreover, we extend the proposed method to handle the scenarios with small samples of raw review data. This study con­ tributes to the literature by exploiting the difference in the distribution of non-fraudulent reviews and that of fraudulent reviews, which can improve the accuracy of fake review detection for online platforms. Collapse

Topics: fake review electronic commerce big data decision support system user behavior

Methods: computational algorithm experiment machine learning semi-supervised learning artificial neural network

Introduction to the Minitrack on Social Commerce: The Good, the Bad, and the Ugly

2023 | HICSS | Citations: 0

Authors: Sundaram, David; Sadovykh, Valeria; Peko, Gabrielle

Abstract: ... Expand

Abstract: Collapse

Topics: electronic commerce technostress mobile application privacy fake review

Methods: sentiment analysis machine learning

Detecting Fake Reviews: Just a Matter of Data

2023 | HICSS | Citations: 0

Authors: Theuerkauf, René; Peters, Ralf

Abstract: Along with the ever-increasing portfolio of products online, the incentive for m ... Expand

Abstract: Along with the ever-increasing portfolio of products online, the incentive for market participants to write fake reviews to gain a competitive edge has increased as well. This article demonstrates the effectiveness of using different combinations of spam detection features to detect fake reviews other than the review-based features typically used. Using a spectrum of feature sets offers greater accuracy in identifying fake reviews than using review-based features only, and using a machine learning algorithm for classification and different amounts of feature sets further elucidates the difference in performance. Results compared by benchmarking show that applying a technique prioritizing feature importance benefits from prioritizing features from multiple feature sets and that creating feature sets based on reviews, reviewers and product data can achieve the greatest accuracy. Collapse

Topics: fake review email spam digital distribution mobile application market mobile application

Methods: machine learning computational algorithm random forest classification term frequency–inverse document frequency Cross-Industry Standard Process for Data Mining

Fake Review Detection - The Value of Domain-Specificity

2023 | Americas Conference on Information Systems | Citations: 0

Authors: Theuerkauf, René; Peters, Ralf

Abstract: The issue of fake reviews, a perennial challenge faced by e-commerce companies, ... Expand

Abstract: The issue of fake reviews, a perennial challenge faced by e-commerce companies, has been worsened in recent years. In this paper, we investigate the trade-off between the application of general-purpose and domain-specific model by selecting training data using transformer-based models to identify fake reviews. Therefore, we compare two scenarios using different set ups of data selection. First, a general-purpose model was identified and applied on specific domains. Afterwards, domain-specific models were trained and tested. Then, the results from these scenarios were compared, yielding the conclusion that models trained on data from a similar product category outperform general-purpose models in classification performance up to 21% (on average by 4%). Our findings send an important message to e-commerce companies to rethink their strategy on training general-purpose models to identify fake reviews to create several, more domainspecific models on that task according to present data domains. Collapse

Topics: fake review online review Facebook electronic commerce recommender system

Methods: RoBERTa DeBERTa BERT transformer model Cross-Industry Standard Process for Data Mining

A deep learning approach for detecting fake reviewers: Exploiting reviewing behavior and textual information

2023 | Decision Support Systems | Citations: 0

Authors: Zhang, Dong; Li, Wenwen; Niu, Baozhuang; Wu, Chong

Abstract: Ensuring the credibility of online consumer reviews (OCRs) is a growing societal ... Expand

Abstract: Ensuring the credibility of online consumer reviews (OCRs) is a growing societal concern. However, the problem of fake reviewers on online platforms significantly influences e-commerce authenticity and consumer trust. Existing studies for fake reviewer detection mainly focus on deriving novel behavioral and linguistic features. These features require extensive human labor and expertise, placing a heavy burden on platforms. Therefore, we propose a novel end-to-end framework to detect fake reviewers based on behavior and textual information. It has two key components: (1) a behavior-sensitive feature extractor that learns the underlying patterns of reviewing behavior; (2) a context-aware attention mechanism that extracts valuable features from online reviews. We rigorously evaluate each proposed module and the entire framework against state-of-the-art benchmarks on two real-world datasets from http://Yelp.com. Experimental results demonstrate that our method achieves state-ofthe-art results on fake reviewer detection. Our method can be considered a tentative step toward lowering human labor costs in realizing automated fake reviewer detection on e-commerce platforms. Collapse

Topics: fake review online review electronic commerce deception detection shapley additive explanation

Methods: machine learning digital trace data experiment parametric test deep learning