Using Stereotypes to Identify Online Transactions
It is not that uncommon to feel uneasy whenever one is shopping in sketchy, questionable sites for a particular product. In particular, there is always a sense of uncertainty in sites that typically are not as well-known as others. The topic of auctions and transactions becomes relevant in addressing this issue as it gives us a sense into what goes into people’s minds when thinking of doing a transaction. Unknown sellers are common in auction sites and buyers simply cannot determine the trustworthiness of these sellers based on their history as it is often not available. This question is explored by Xin Liu, et. al in a study aiming to use stereotypes to identify risky transactions in internet auctions. The study aims to evaluate the strength of stereotypes by identifying discriminating attributes from local knowledge and community knowledge to distinguish successful and unsuccessful transactions. As it pertains to the mechanics of the study, the researchers use data from Allegro, an auction site in Eastern Europe.
The researchers begin this study by setting thresholds in an effort to predict frauds through low false positive scores and successful transactions through low false negative scores. Determining the authenticity of a seller and trustworthiness solely through individual input is virtually impossible, hence the researches aim to explore the extent to which community knowledge can assist the evaluation of a given transaction. The impact of community knowledge becomes important in deriving stereotypes “for inexperienced buyers that dominate online auction sites” (Liu, et. al). The researchers conclude that it is indeed possible to determine legitimate transactions by building stereotypes that aggregate past experiences or “cumulating the experience from the stereotypes to which the potential transaction conforms” (Liu, et. al). False positives and false negatives become particularly important in evaluating this question while also taking into account how much knowledge there is available from the individual and the community. Determining the legitimacy of a transaction, thus, can be ameliorated by taking into account stereotypes that conform to the transaction from the individual’s past experience. Alternatively, that is if individual knowledge on the matter is lacking, community knowledge should be taken into account to aid this decision. Although this approach is by no means the most ethical one, it goes to show how much biases can impact online auctions in both negative and positive ways.
Source: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5591325