Barnaby Page who moderated several sessions at DS Asia 2007 so well made the quip that a DS conference these days wasn’t complete until someone had showed a clip from the film “Minority Report” – have you seen how many “Minority Report” clips there are on YouTube?
Anyway, maybe today’s modern blog isn’t complete without a mention either – although the thesis discussed below works more along the lines of RFID than eye recognition systems.
Imran Ashraf from Rotterdam University is doing a thesis on what can be achieved if you combine: –
- Digital Signage
- Item-level RFID-tagging
- Recommendation systems (data mining on past purchases to determine what a customer might be interested in; “people who bought this book also bought this book”)
The idea being a screen that gives recommendations for other products based on what you hold in front of it!!
The final part of the research consists of a survey which Imran is asking people to spend a few minutes filling in.
Imran explains At the moment item-level RFID tagging is not widely used, because the RFID-chips are expensive compared to barcode. But eventually it will become mainstream. The idea is that when you walk past the screen with a product that you want to buy, that the screen automatically detects which product you are holding in your hand. It scans the RFID-chip and looks up in the store-database what kind of product it is. The recommender system uses that information to determine what kind of products you might also be interested in as well and shows those on the screen. It is a way to increase cross-sells. At the moment RFID is used mainly in the supply chain, but eventually it will also be used to tag each object individually (to follow each object in the chain). The idea that I am proposing is using this for something what it was never intended for by using existing technologies. In my thesis I look at all these 3 technologies, combine them together and perform a economic (ROI, etc) and technical analysis.
Results will be shared I am told – just leave your email address at the end of the survey.
The thesis will be finished by Jan 2008 and we will be writing about it more here in the New Year.