Recommender system for online dating service
To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Copyright 2012 ACM 978-1-4503-1638-5/12/09 ....00. DISSIMILAR LIKE DISLIKE SIMILAR Figure 1: The four kinds of relationships modeled in our approach: Like, dislike, similarity and dissim- ilarity.
metric and clearly divided into those who receive a recom- mendation (the user) and that what is recommended (the product).
Between the power of suggestion, knowledge of users’ tastes, and lack of barriers between hitting “purchase” and having the treat delivered – what role should ethically-responsible retailers play in helping their users avoid decisions that could negatively impact their well-being?
Unfortunately, there’s unlikely to be a one-size-fits-all approach across sectors and systems.
We show that this unified representation is capable of modeling both notions of relations between users in a joint expression and apply it for recommending potential partners.
In exper- iments with the Czech dating website we show that our modeling approach leads to an improvement over baseline recommendation methods in this scenario.
Fortunately, research and methods are underway to redirect and limit radicalizing behavior.
Here, a user receives suggestions for people he might know based on his social context.
In this case the setting is symmetric, as there is no distinction between the type of the object that is recommended and the one that receives the recommendation (both are users).
It doesn’t take long to think up recommendation scenarios that could raise an eyebrow.
While humans fold ethical considerations into their recommendations, algorithms programmed to drive revenue do not.
However, dangers need not be as extreme as ISIS sympathizing to merit notice. With over 70% of the US population projected to make an online purchase this year, behind-the-scenes algorithms could be influencing the purchasing habits of a healthy majority of the population.