Counterfeit product detection: Bridging the gap between design science and behavioral science in information systems research

Highlights

• This paper presents a design artifact, called OnCDS, to assist consumers in detecting counterfeit products.

• We unite design and behavioral science via a design artifact and behavioral model using kernel theory and valence framework.

• OnCDS uses web scraping, NLP, and topic analysis to calculate the likelihood of a product being counterfeit.

• We evaluate its utility in terms of accuracy of counterfeit scores and their effect on a consumer’s attitude.

• Results show that OnCDS’ efficacy is validated and that the counterfeit score affects perceived risk, trust, and attitude.

Abstract

In IS research, there is a dichotomy where design science and behavioral science are distinct research paradigms. IS researchers should view these paradigms as complementary with research drawing upon the strengths of both, yet few have done so. This work demonstrates how design science and behavioral science can be united in IS research via counterfeit product detection based on product reviews in an online marketplace. Product authenticity in the online marketplace is a common issue plaguing consumers. The decision process involved in determining product authenticity is lengthy and complex. Despite the pressing need for an automatic authenticity rating system for online shopping, little research has been done to develop such a system and assess its effects on consumer purchase behavior. To respond to this need, our study develops a design artifact, called OnCDS, to automatically calculate the likelihood that a product is counterfeit based on online customer reviews. Drawing upon lexicon-based sentiment analysis approaches and TF-IDF as kernel theories for our design, we employ web scraping, natural language processing, and topic analysis methods to process customer reviews and calculate the counterfeit score of a product. In assessing the effects of OnCDS on consumer behavior, we develop a research model that encompasses trust and perceived risk based on the valence framework. Results show that our design artifact's efficacy is validated and that the counterfeit score affects perceived risk and trust, which in turn influences attitude toward purchase.

Introduction

Counterfeit goods, specifically luxury goods, comprise a rapidly expanding industry [2]. Counterfeiting has expanded and flourished in part due to the demand for luxury goods [12]. Pervasive counterfeit goods in the marketplace devalue product branding and have a significant negative impact on the global economy. According to estimates from the Organization for Economic Co-operation and Development (OECD), international trade on counterfeit and pirated goods accounted for USD 461 billion in 2013 [67], which is approximately 2.5% of world trade. These estimates unfortunately represent a substantial increase from those in 2008. A similar OECD study in 2008 reported that counterfeit and pirated products were worth USD 200 billion, which is about 1.9% of world trade. The 2009 report created by the organization Business Action to Stop Counterfeiting and Piracy (BASCAP) estimates that USD 62 billion is lost annually from tax revenue due to counterfeiting and piracy. Moreover, the BASCAP report also states that there is a USD 20 billion increase in related crimes and a USD 14.5 billion in lost lives cost, all of which is augmented by USD 100 million in additional healthcare services required for fake products (e.g. medications) [5]. Compounding the issue is the fact that purchasers of counterfeit goods often feel the goods are comparable to authentic brand goods and are unaware of the potential negative effects that purchasing counterfeit products has on the economy [66].

Online marketplaces, such as Amazon.com and eBay, have low barriers to entry and thus are more vulnerable for counterfeit products than traditional brick and mortar establishments. Fraudsters can easily assume different or multiple identities, making them difficult to trace [19]. The advent of technology has produced conditions which can be exploited by criminals [64]. Further, counterfeit products are frequently sold via legitimate websites [93]. One such example is the premium brand Tiffany & Co. The company launched, and subsequently lost, a lawsuit against eBay due to the large amount of its counterfeit merchandise on eBay's online auction website [39]. Tiffany & Co. had employees purchase 325 items from eBay listed as Tiffany & Co. products and determined that 75% of the items were counterfeit [14]. EBay implemented buyer protections against counterfeit products; however, they are still exchanged on a worldwide basis. Amazon is not only a seller but has evolved into an online marketplace that brings sellers from across the world together on the Amazon platform to sell their products alongside Amazon's products on Amazon.com [3]. According to Martin [61] of Entrepreneur.com, Amazon is not just another website but the king of e-commerce platforms where sellers can easily enter the market and sell their products. On the downside, in 2017, the number of scams on Amazon.com increased, fraudulent sellers were prevalent, and fraudulent sellers absconded with massive profits [82].

Counterfeit goods in the online marketplace are a serious issue plaguing consumers. The decision-making process involved in determining the authenticity of a product is lengthy and complex. A consumer must review a large volume of qualitative information, such as product and seller reviews, to determine the authenticity of a product. Sellers can easily enter and leave the online marketplace, thereby adding factors into an already complex decision process. Web crawlers, application programming interfaces, natural language processing, and topic analysis may be employed to support the consumer's decision-making process by automatically rating the authenticity of products in online marketplaces, such as Amazon. Further, authenticity ratings may improve trust in the product, thereby increasing its sales. While literature presents abundant research on e-commerce, they focus largely on consumer behavior and purchasing counterfeit products, remaining in the realm of behavioral research. Unfortunately, a dichotomy exists within IS research with design science and behavioral research viewed as distinct research paradigms. IS researchers should view these paradigms as complementary with research drawing upon the strengths of both. Despite the usefulness and potential benefits of an automatic authenticity rating system in online shopping, few research has proposed such a system and empirically evaluated its effects on consumer purchase behavior, which bridges the two IS research paradigms.

To fill in this gap, this study develops an automatic counterfeit scoring system, called Online Counterfeit Detection Score (OnCDS), to support the consumer's decision making process by identifying counterfeit goods based on consumer product reviews and empirically tests its utility using a research model drawn upon the valence framework [76] in consumer behavior. The design science research methods [46], [74] and information systems design theory [36] provide the guidelines for designing OnCDS. Adopting lexicon-based sentiment analysis and Term Frequency/Inverse Document Frequency (TF-IDF) as kernel theories for our study, we employ web scraping, natural language processing, and topic analysis to process customer reviews from an online marketplace and calculate the counterfeit scores of products. We instantiate our design artifact and evaluate its performance with human subjects recruited from Amazon Mechanical Turk. Additionally, we empirically test our research model that hypothesizes the effect of our design artifact on consumer purchase attitude. Results show that the counterfeit score affects trust and perceived risks, which in turn affect attitude toward purchase. Few prior studies have performed behavioral research to assess the utilities of a proposed design artifact. As a result, this study significantly contributes to IS literature by not only developing an automatic counterfeit scoring system based on consumer product reviews, but also by demonstrating how two IS research paradigms, design science and behavioral research, can be united to raise research rigor and relevance.

The remainder of this paper is organized as follows. Section 2 provides the literature review, followed by the discussion on the dichotomy in IS research in Section 3. Section 4 presents the theoretical framework and research hypotheses. Section 5 describes our design artifact, OnCDS, its implementation details, and its performance evaluation. Section 6 explains our behavioral research methods to empirically test the effects of our proposed artifact on consumer behavior and the test results. Section 7 presents the discussion of study implications and offers conclusions and future research directions.