Your task is to analyze an existing recommender system that you find interesting. You should:
Perform a Scenario Design analysis as described below. Consider whether it makes sense for your selected recommender system to perform scenario design twice, once for the organization (e.g. Amazon.com) and once for the organization’s customers.
Attempt to reverse engineer what you can about the site, from the site interface and any available information that you can find on the Internet or elsewhere.
Include specific recommendations about how to improve the site’s recommendation capabilities going forward.
Create your report using an R Markdown file, and create a discussion thread with a link to the GitHub repo where your Markdown file notebook resides. You are not expected to need to write code for this discussion assignment.
Hinge is among one of the most popular dating apps available. Unlike Bumble, Tinder, Match, and others of the same, Hinge has a more immersive UI experience and claims that “it is meant to be deleted”. Their algorithm is inspired by the \(Stable Marriage Algorithm\), by David Gale and Lloyd Shappley. Their experiment proved that if there’s an equal amount of men and women, then there can be enough combinations for everyone. \(^1\)
As far as Hinge is concerned, this translates to everyone ending up with someone they like the most. However, this is not possible without each user filling out the “personal” and “preference” profiles. Matches are made by comparing the most compatible responses together. If someone wishes to be with someone who is within a certain height, has a certain orientation, or has attained a certain education level, then Hinge will show those users their profiles. On the converse, if one absolutely does not want to be with someone who has a subjectively disaproving lifestyle, then they won’t be seen either.
In 2018, it was reported that 75% of first dates on Hinge turned into second dates. On the first date, 90% of users said it was great, and 72% expressed interest to go on a second. Also, if one has the word “\(chipotle\)” in their profile and menu items, then they were 2.4x more likely to speak with another user. Apparently, if users included the menu options “queso” and “chorizo”, this lead to 70% more likes and to the most dates respectively. The target users of Hinge are mostly millenials: 90% are bewteen the ages of 23 and 36, with 99% being college educated. Women comprise 50.2%, and men 49.8%. \(^2\)
As far as implementing a scenario analysis, it does not seem necessary. Their population differences are almost exactly what’s needed for their reccomendation system to work in order to maximize the \(Stable Marraige Algorithm\). However, finding ways to improve the functionality of machine learning qould improve the reccomendation system. But this delves more into processing and computation power, or advanced ML algorithms–all of which I don’t know much of.