Deceptive Opinion Research Articles

A Weighted Stacking Ensemble Model With Sampling for Fake Reviews Detection

Customers use reviews as a primary source of information to judge a product or service. Positive reviews help boost companies’ reputations, increasing their revenue by attracting new clients, and increasing the purchasing order size. On the other hand, negative reviews significantly reduce sales, which might be the case due to competitive advantage. Organizations can use fake (i.e., misleading or fraudulent) reviews to generate fast profits by deceiving customers into buying their products. Recently, various methods to assess the legitimacy of reviews have been introduced using advances in machine learning. However, existing methods fall short of achieving highly accurate detection results for unbalanced classes. We aimed to create a spam review identification model using ensemble-based learning while balancing classes using sampling techniques. This article proposes a weighted stacking ensemble model with sampling (WSEM-S) for efficient fake reviews detection. We used <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$n$</tex-math> </inline-formula> -gram models to effectively model language data for feature retrieval. The experimental results on three customer reviews datasets: YELPNYC, Deceptive Opinion Spam Corpus (DOSC) v1.4, and Deception datasets show that the proposed model outperforms the conventional machine learning techniques Naïve Bayes, logistic regression, K-nearest neighbor (KNN), random forest, extreme gradient boosting (XGBoost), and convolutional neural network (CNN) as well as the state-of-the-art ensemble models.

A Contextual Relationship Model for Deceptive Opinion Spam Detection.

The promotion of e-commerce platforms has changed the lifestyle of several people from traditional marketing to digital marketing where businesses are made online and the concurrence reached high levels. These platforms have helped the ease of purchases while providing more advantages to the customers such as benefiting from a wide range of high-quality products, low prices, buying at any time, and more importantly supplying information and reviews about the products, and so on. Unfortunately, a plethora of companies mislead the customers to buy their products or demote the competitors' by using deceptive opinion spams which has a negative impact on the decision and the behavior of the purchasers. Deceptive opinion spams are written deliberately to seem legitimate and authentic so that to misguide or delude the customer's purchases. Consequently, the detection of these opinions is a hard task due to their nature for both humans and machines. Most of the studies are based on traditional machine learning and sparse feature engineering. However, these models do not capture the semantic aspect of reviews. According to many researchers, it is the key to the detection of deceptive opinion spam. Besides, only a few studies consider using contextual information by adopting neural networks in comparison with plenty of traditional machine learning classifiers. These models face numerous shortcomings as long as their representations are obtained while mining each review considering only words, sentences, reviews, or a combination of them, thereby classifying them based on their representations. In fact, deceptive opinions are written by the same deceivers belonging to the same companies with similar aims to promote or demolish a product. In other words, Deceptive opinion spams tend to be semantically coherent with each other. To the best of our knowledge, no model tries to obtain a representation based on the contextual relationships between opinions. This article proposes to use a capsule neural network, bidirectional long short-term memory, attention mechanism, and paragraph vector distributed bag of words to detect deceptive opinion spam. Our model provides a powerful representation of the opinions since it centers on the preservation of their contexts and the relationships between them. The results show that our model significantly outperforms the existing state-of-the-art models.

Interviewing to Detect Lies About Opinions: The Devil’s Advocate Approach

The verbal cues lie tellers spontaneously report during interviews are weak and unreliable (DePaulo et al., 20023; Vrij, Hartwig, &amp; Granhag, 2019). Researchers therefore asked for interview protocols to be developed that would elicit such verbal cues (Vrij &amp; Granhag, 2012). Ten years later several interview protocols have emerged (Vrij, Granhag et al., 2022), including Cognitive Credibility Assessment (Vrij, Fisher, &amp; Blank, 2017; Vrij, Mann et al., 2021), Reality Interviewing (Bogaard et el., 2019; Colwell et al., 2007), the Strategic Use of Evidence (Granhag &amp; Harwig, 2015; Hartwig et al., 2014) and the Verifiability Approach (Nahari, 2019; Palena et al., 2021). All these interview protocols focus on distinguishing between truthful and deceptive statements about alleged activities. However, practitioners are also interested in detecting lies when people discuss their opinions. The Devil’s Advocate Approach is the only interview protocol we are aware of aimed to distinguish between truthful and deceptive opinions. In this article we present the (scarce) Devil’s Advocate Approach research conducted to date.

Deceptive opinion spam detection using feature reduction techniques

Deceptive opinion spam detection using feature reduction techniques

DETECTING DECEPTIVE REVIEWS: AN INTEGRATED MACHINE LEARNING APPROACH

In recent years, online reviews have become a crucial factor in promoting products and services. However, the rise of fake reviews has posed a significant challenge. Businesses, marketers, and advertisers often resort to embedding fake reviews to attract customers or undermine their competitors. Deceptive reviews have become a common practice, as they serve as a means of promoting one''''s own business or tarnishing the reputation of rivals. Consequently, the identification of deceptive reviews has emerged as a critical and ongoing research area. This research paper presents a machine learning model approach to detect deceptive reviews. The study focuses on experiments conducted using a deceptive opinion spam corpus dataset, specifically targeting restaurant reviews. An n-gram model combined with max features is developed to identify deceptive content, with a particular emphasis on fake reviews. Additionally, a benchmark study is conducted to explore the performance of two different feature extraction techniques and their application in five machine learning classification techniques. The experimental findings demonstrate that the passive aggressive classifier outperforms other algorithms, achieving the highest accuracy not only in text classification but also in identifying fake reviews. Moreover, the research delves into the identification of deceptive reviews and explores diverse feature extraction and machine learning techniques to improve the model''''s accuracy.

Leveraging Readability and Sentiment in Spam Review Filtering Using Transformer Models

Online reviews significantly influence decision-making in many aspects of society. The integrity of internet evaluations is crucial for both consumers and vendors. This concern necessitates the development of effective fake review detection techniques. The goal of this study is to identify fraudulent text reviews. A comparison is made on shill reviews vs. genuine reviews over sentiment and readability features using semi-supervised language processing methods with a labeled and balanced Deceptive Opinion dataset. We analyze textual features accessible in internet reviews by merging sentiment mining approaches with readability. Overall, the research improves fake review screening by using various transformer models such as Bidirectional Encoder Representation from Transformers (BERT), Robustly Optimized BERT (Roberta), XLNET (Transformer-XL) and XLM-Roberta (Cross-lingual Language model–Roberta). This proposed research extracts and classifies features from product reviews to increase the effectiveness of review filtering. As evidenced by the investigation, the application of transformer models improves the performance of spam review filtering when related to existing machine learning and deep learning models.

Deceptive reviews and sentiment polarity: Effective link by exploiting BERT

Today, reviews are the advertising medium par excellence through which companies are able to influence customers’ spending decisions. Although the initial purpose of reviews was to provide companies with a feedback tool to improve products and services based on customer needs, they soon became a way to climb the sales rankings, often illegally. In fact, deceptive and fake reviews have managed to evade the often non-existent means of validation of online platforms, proliferating a new business. To combat this phenomenon, several classification methods have been developed to train automated tools in the arduous task of distinguishing between genuine and misleading reviews, the most recent based on machine and deep learning techniques. This paper proposes a multi-label classification methodology based on the Google BERT neural language model to build a deceptive review detector aided by its sentiment awareness: improved modeling of the link between sentiment polarity and deceptiveness during the fine-tuning phase by exploiting the Binary Cross Entropy with Logits loss function adds to the advantages provided by pre-trained contextual models, which are able to capture word polysemy through word embeddings and benefit from pre-training on huge corpora. Tests were performed on the Deceptive Opinion Spam Corpus and Yelp New York City datasets, providing a quantitative and qualitative analysis of the results which, when compared with the state of the art available in the literature, showed an encouraging increase in performance.

Deceptive opinion spam detection approaches: a literature survey

Deceptive opinion spam detection approaches: a literature survey

Data Analytics for the Identification of Fake Reviews Using Supervised Learning

Fake reviews, also known as deceptive opinions, are used to mislead people and have gained more importance recently. This is due to the rapid increase in online marketing transactions, such as selling and purchasing. E-commerce provides a facility for customers to post reviews and comment about the product or service when purchased. New customers usually go through the posted reviews or comments on the website before making a purchase decision. However, the current challenge is how new individuals can distinguish truthful reviews from fake ones, which later deceives customers, inflicts losses, and tarnishes the reputation of companies. The present paper attempts to develop an intelligent system that can detect fake reviews on e-commerce platforms using n-grams of the review text and sentiment scores given by the reviewer. The proposed methodology adopted in this study used a standard fake hotel review dataset for experimenting and data preprocessing methods and a term frequency-Inverse document frequency (TF-IDF) approach for extracting features and their representation. For detection and classification, n-grams of review texts were inputted into the constructed models to be classified as fake or truthful. However, the experiments were carried out using four different supervised machine-learning techniques and were trained and tested on a dataset collected from the Trip Advisor website. The classification results of these experiments showed that naïve Bayes (NB), support vector machine (SVM), adaptive boosting (AB), and random forest (RF) received 88%, 93%, 94%, and 95%, respectively, based on testing accuracy and tje F1-score. The obtained results were compared with existing works that used the same dataset, and the proposed methods outperformed the comparable methods in terms of accuracy.

What Yelp Fake Review Filter Might Be Doing?

Online reviews have become a valuable resource for decision making. However, its usefulness brings forth a curse ‒ deceptive opinion spam. In recent years, fake review detection has attracted significant attention. However, most review sites still do not publicly filter fake reviews. Yelp is an exception which has been filtering reviews over the past few years. However, Yelp’s algorithm is trade secret. In this work, we attempt to find out what Yelp might be doing by analyzing its filtered reviews. The results will be useful to other review hosting sites in their filtering effort. There are two main approaches to filtering: supervised and unsupervised learning. In terms of features used, there are also roughly two types: linguistic features and behavioral features. In this work, we will take a supervised approach as we can make use of Yelp’s filtered reviews for training. Existing approaches based on supervised learning are all based on pseudo fake reviews rather than fake reviews filtered by a commercial Web site. Recently, supervised learning using linguistic n-gram features has been shown to perform extremely well (attaining around 90% accuracy) in detecting crowdsourced fake reviews generated using Amazon Mechanical Turk (AMT). We put these existing research methods to the test and evaluate performance on the real-life Yelp data. To our surprise, the behavioral features perform very well, but the linguistic features are not as effective. To investigate, a novel information theoretic analysis is proposed to uncover the precise psycholinguistic difference between AMT reviews and Yelp reviews (crowdsourced vs. commercial fake reviews). We find something quite interesting. This analysis and experimental results allow us to postulate that Yelp’s filtering is reasonable and its filtering algorithm seems to be correlated with abnormal spamming behaviors.

Fast Detection of Deceptive Reviews by Combining the Time Series and Machine Learning

With the rapid growth of online product reviews, many users refer to others’ opinions before deciding to purchase any product. However, unfortunately, this fact has promoted the constant use of fake reviews, resulting in many wrong purchase decisions. The effective identification of deceptive reviews becomes a crucial yet challenging task in this research field. The existing supervised learning methods require a large number of labeled examples of deceptive and truthful opinions by domain experts, while the available unsupervised learning methods are inefficient because they depend on the features of reviewers to detect each fake review. Therefore, by focusing on the detection efficiency problem and the limitation of large amount of labeled examples dependence, in this paper, we proposed an effective semisupervised learning approach for detecting spam reviews. Firstly, a time series model of all the reviews of a product is constructed, and then the suspected time intervals are captured based on the burst review increases in these intervals. Secondly, a co-training two-view semisupervised learning algorithm was performed in each captured interval, in which linguistic cues, metadata, and user purchase behaviors were synthetically employed to classify the reviews and check whether they are spam ones or not. A series of numerical experiments on a real dataset acquired from Taobao.com have confirmed the effectiveness of the proposed model, not only reaping benefits in terms of time efficiency and high accuracy but also overcoming the shortcomings of supervised learning methods, which depend on large amounts of labeled examples. And a trade-off balance was obtained between accuracy and efficiency.

Deceptive Opinions Detection Using New Proposed Arabic Semantic Features

Some users try to post false reviews to promote or to devalue other’s products and services. This action is known as deceptive opinions spam, where spammers try to gain or to profit from posting untruthful reviews. Therefore, we conducted this work to develop and to implement new semantic features to improve the Arabic deception detection. These features were inspired from the study of discourse parse and the rhetoric relations in Arabic. Looking to the importance of the phrase unit in the Arabic language and the grammatical studies, we have analyzed and selected the most used unit markers and relations to calculate the proposed features. These last were used basically to represent the reviews texts in the classification phase. Thus, the most accurate classification technique used in this area which has been proven by several previous works is the Support Vector Machine classifier (SVM). But there is always a lack concerning the Arabic annotated resources specially for deception detection area as it is considered new research area. Therefore, we used the semi supervised SVM to overcome this problem by using the unlabeled data.

A non-convex semi-supervised approach to opinion spam detection by ramp-one class SVM

Nowadays, e-commerce has become a part of our daily life in such a way that people's decision for buying products or choosing services highly depends on comments, reviews, and rates, which are posted on related businesses’ website and other social media. Because of the importance and prevalence of these sources of information, fraudsters are tempted to use fraudulently opinion sharing platforms in order to promote or to discredit some target products or services. Although a wide range of approaches have been proposed to address this problem and to help distinguish between the deceptive or fraudulent opinions from the trustful ones, this is still a challenging problem. Lack of well-defined deceptive data samples or insufficiency of spam review instances in the training sets cause supervised techniques facing an imbalanced classes problem. Furthermore, even in the golden normal opinion datasets, there is a possibility of the presence of some abnormal records or outliers. To deal with these two issues, we propose a robust and non-convex semi-supervised algorithm called “Ramp One-Class SVM”. In the proposed method, one-class SVM is adopted to handle the lack of labeled data for the deceptive opinions and by taking the advantages of non-convex properties of the Ramp loss function, we eliminate the effects of outliers and non-review opinions. The performance of the proposed method is evaluated by an artificial dataset and two real datasets including Ott and Yelp crowdsourced datasets. The results show the superiority of our method by achieving an accuracy of 92.13% and 74.37% for Ott and Yelp crowdsourced datasets, respectively. The obtained results also reveal the effectiveness of the proposed model in terms of precision, recall, generalization power, and robustness to the outliers.

PV-DAE: A hybrid model for deceptive opinion spam based on neural network architectures

Opinion review is of great importance for both customers and organizations. Indeed, it helps customers in buying decisions and represents a valuable feedback for the companies, allowing them to improve their productions. However, numerous greedy companies resort to fake reviews in order to influence the customer and brighten the brand image, or to defame the one of their competitors. Various models are proposed in order to detect deceptive opinion reviews. Most of these models adopt traditional methods focusing on feature extraction and traditional classifiers. Unfortunately, these models do not capture the semantic aspect while ignoring the opinion’s context. In order to tackle this issue, we propose a new approach based on Paragraph Vector Distributed Bag of Words (PV-DBOW) and the Denoising Autoencoder (DAE). The proposed customized model provides a strong representation which is based on a global representation of the opinions while preserving their semantics. Indeed, the embedding vectors capture the semantic meaning of all words in the context of each opinion. The generated review representations are fed into a fully connected neural network in order to detect deceptive opinion spam. The obtained results concerning the deception dataset show that our model is effective and outperforms the existing state-of-the-art methodologies.

Identification of Opinion Spammers using Reviewer Reputation and Clustering Analysis

Online reviews have increasingly become a very important resource before making a purchasing decisions. Unfortunately, malicious sellers try to game the system by hiring a person or team (which is called spammers) to fabricate fake reviews to improve their reputation.Existing methods mainly take the problem as a general binary classification or focus on some heuristic rules. However, supervised learning methods relies heavily on a large number of labeled examples of deceptive and truthful opinions by domain experts, and most of features mentioned in the heuristic strategy ignore the characteristic of the group organization among spammers. In this paper, an effective method of identifying opinion spammers is proposed. Firstly, suspected spammers are detected by means of unsupervised learning based on reviewer’s reputation. We believe that the reviewer’s reputation has a direct relation with the quality of reviews. Generally, review written by user with lower reputation, shows lower quality and higher possibility to be fake. Therefore, the model assigns reputation score to each reviewer wherein the content based factors and activeness of reviewers are employed efficiently. On basis of all suspected spammers, k-center clustering algorithm is performed to further spot the spammers based on the observation of burst of review release time. Experimental results on Amazon’s dataset are encouraging and indicate that our approach poses high accuracy and recall, and good performance is achieved.

Robustness of Word and Character N-gram Combinations in Detecting Deceptive and Truthful Opinions

Opinions in reviews about the quality of products or services can be important information for readers. Unfortunately, such opinions may include deceptive ones posted for some business reasons. To keep the opinions as a valuable and trusted source of information, we propose an approach to detecting deceptive and truthful opinions. Specifically, we explore the use of word and character n-gram combinations, function words, and word syntactic n-grams (word sn-grams) as features for classifiers to deal with this task. We also consider applying word correction to our utilized dataset. Our experiments show that classification results of using the word and character n-gram combination features could perform better than those of employing other features. Although the experiments indicate that applying the word correction might be insignificant, we note that the deceptive opinions tend to have a smaller number of error words than the truthful ones. To examine robustness of our features, we then perform cross-classification tests. Our latter experiments results suggest that using the word and character n-gram combination features could work well in detecting deceptive and truthful opinions. Interestingly, the latter experimental results also indicate that using the word sn-grams as combination features could give good performance.

Machine Cleaning of Online Opinion Spam: Developing a Machine-Learning Algorithm for Detecting Deceptive Comments

Humans are not very good at detecting deception. The problem is that there is currently no other particular way to distinguish fake opinions in a comments section than by resorting to poor human judgments. For years, most scholarly and industrial efforts have been directed at detecting fake consumer reviews of products or services. A technique for identifying deceptive opinions on social issues is largely underexplored and undeveloped. Inspired by the need for a reliable deceptive comment detection method, this study aims to develop an automated machine-learning technique capable of determining opinion trustworthiness in a comment section. In the process, we have created the first large-scale ground truth dataset consisting of 866 truthful and 869 deceptive comments on social issues. This is also one of the first attempts to detect comment deception in Asian languages (in Korean, specifically). The proposed machine-learning technique achieves nearly 81% accuracy in detecting untruthful opinions about social issues. This performance is quite consistent across issues and well beyond that of human judges.

Learning to Detect Deceptive Opinion Spam: A Survey

With the development of e-commerce, more and more users begin to post reviews or comments about the quality of products on the internet. Meanwhile, people usually read previous reviews before purchasing online products. However, people are frequently deceived by deceptive opinion spam, which is usually used for promoting the products or damaging their reputations because of economic benefit. Deceptive opinion spam can mislead people's purchase behavior, so the techniques of detecting deceptive opinion spam have extensively been researched in past ten years. In particular, some work based on deep learning has been investigated in last three years for the task. However, there still lack a survey, which can systematically analyze and summarize the previous techniques. To address this issue, this paper first introduces the task of deceptive opinion spam detection. Then, we summarize the existing dataset resources and their construction methods. Third, existing methods are analyzed from two aspects: traditional statistical methods and neural network models. Finally, we give some future directions of the task.

Towards Accurate Deceptive Opinions Detection Based on Word Order-Preserving CNN

Convolutional neural network (CNN) has revolutionized the field of natural language processing, which is considerably efficient at semantics analysis that underlies difficult natural language processing problems in a variety of domains. The deceptive opinion detection is an important application of the existing CNN models. The detection mechanism based on CNN models has better self-adaptability and can effectively identify all kinds of deceptive opinions. Online opinions are quite short, varying in their types and content. In order to effectively identify deceptive opinions, we need to comprehensively study the characteristics of deceptive opinions and explore novel characteristics besides the textual semantics and emotional polarity that have been widely used in text analysis. In this paper, we optimize the convolutional neural network model by embedding the word order characteristics in its convolution layer and pooling layer, which makes convolutional neural network more suitable for short text classification and deceptive opinions detection. The TensorFlow-based experiments demonstrate that the proposed detection mechanism achieves more accurate deceptive opinion detection results.

DRI-RCNN: An approach to deceptive review identification using recurrent convolutional neural network

With the widespread of deceptive opinions in the Internet, how to identify online deceptive reviews automatically has become an attractive topic in research field. Traditional methods concentrate on extracting different features from online reviews and training machine learning classifiers to produce models to decide whether an incoming review is deceptive or not. This paper proposes an approach called DRI-RCNN (Deceptive Review Identification by Recurrent Convolutional Neural Network) to identify deceptive reviews by using word contexts and deep learning. The basic idea is that since deceptive reviews and truthful reviews are written by writers without and with real experience respectively, the writers of the reviews should have different contextual knowledge on their target objectives under description. In order to differentiate the deceptive and truthful contextual knowledge embodied in the online reviews, we represent each word in a review with six components as a recurrent convolutional vector. The first and second components are two numerical word vectors derived from training deceptive and truthful reviews, respectively. The third and fourth components are left neighboring deceptive and truthful context vectors derived by training a recurrent convolutional neural network on context vectors and word vectors of left words. The fifth and six components are right neighboring deceptive and truthful context vectors of right words. Further, we employ max-pooling and ReLU (Rectified Linear Unit) filter to transfer recurrent convolutional vectors of words in a review to a review vector by extracting positive maximum feature elements in recurrent convolutional vectors of words in the review. Experiment results on the spam dataset and the deception dataset demonstrate that the proposed DRI-RCNN approach outperforms the state-of-the-art techniques in deceptive review identification.