Solving difficult problems using play and collective intelligence

The Problem
This problem came apparent when trying to design a “Gift Recommendation service”. The general idea is to use a retailers API and your contacts interests (interest graphs built using LIKES, Follows, Favourites using data available via the users social networks – FB, Twitter, LinkedIn) to recommend relevant gifts (based on their interests, occasion, and strength of the relationship). This was the easy part – the difficulty came when trying to associate the contacts ‘interests’ with ‘gifts (products). One way would base it on keywords e.g. filter products using features from their description and the contacts interests (‘sport’, ‘music’, …) – I imagine this approach would have failed miserably.

The Opportunity 

  • Ubiquity of Smartphones
  • Forming habits (users unlocking their phone, on average, 160 times per day)
  • Rise of Micro-interactions
  • Rise of casual gaming

A Solution
I remember watching a game show once where they had partners (or friends/family – was a while ago) on – they kept one behind the stage and the other on stage. The game was to determine how well they knew each other by asking a series of questions about personal taste & subjective points of view e.g. Favourite movie (multi-chocie). Because of the ubiquity of Smartphone devices, micro-etc – it wouldn’t be difficult to adopt this concept to problems similar to the one I have e.g. the game would be ‘knowing your friends’ – the objective is to match the products that he/she would like (with the peer setting the correct answer). If designed and marketed well you should have no problem building a large enough dataset for classification (based on interests and classified product/product types).

Currently still just my theory – the aim is to implement this later on this year to better understand the opportunity of crowd sourcing using play. Get in contact if this is something of interest/want to get involved or know of similar examples.