PUPS: Personalised Ubiservices in Public Spaces
The project designs, implements and field trials prototypes of novel ubicomp applications that combine context-awareness with personalisation in order to provide a better user experience in everyday tasks in public spaces. In contrast to most of earlier research, our focus is outside the home and workplace. Our showcase is the Itäkeskus shopping mall, but the research results can also be applied to other public spaces. Our application areas are consumer applications and social mobile media. We aim at ideas that could be commercialised in about five years time.
Our information flyer about the project includes further information. Our main concepts are the Ma$$iv€ intelligent shopping assitant and the Funnelry social media aggregator.
Our project presentation at the Ubicom results seminar on 25 September 2009 is here; we also have a number of videos:
- Ma$$iv€ - intelligent shopping assistant
- Funnelry- combining social media
- UbiLight - interaction by moving the hand
Our project presentation in Japan in December 2008 is here.
We have made a small video about a virtual post-it note application we made. Further videos are in production.
Link to a questionnaire about shopping lists and preferences of Ma$$iv€ features.
The following scenario inspires the project and can be seen as a long-term goal for the project:
Leena intends to buy food for her family for the next two days. Before leaving to the store, she makes a shopping list on her mobile device. Her operator provides a service called Ma$$iv€. This observes from the list of items she is entering, that she apparently is going to make meat balls. However, she has not entered onions, so Ma$$iv€ suggests: “Have you possibly forgotten the onions?” Ma$$iv€ also notes that Leena’s family usually eats cornflakes at breakfast. Cornflakes are now on promotion, so Ma$$iv€ informs Leena about this.
When arriving at the mall, Leena decides to go first to the cafeteria. When seated, she opens the Funnelry social mobile media application to check her friends’ recent and current doings. The application displays a condensed view of what her contacts in IRC-galleria, Flickr and Blogger, etc. have submitted there recently. She sees that her friend Kirsi has just recently shot a video clip of herself at the mall and submitted that to YouTube. When Leena opens the link to view the video, Kirsi gets a notification on her mobile device. She has selected to be notified immediately whenever some of her friends is looking at or commenting her videos. Kirsi immediately checks Funnelry whether there are any hints of Leena’s location in either Plazes, Jaiku or Imity. When Leena is still looking at the video, Kirsi contacts her via IM, asking what she is up to at the mall. Leena does not feel like typing right now, and replies via MMS sending only a photo of the muffin she is eating.
When entering the shop, the shop’s system recognises Leena and provides additional services over a voice interface. On the basis of the shopping list, the system suggests an optimised itinerary in the shop and orders the shopping list items in the order they appear on this itinerary. When Leena accidentally skips an item, Ma$$iv€ notes that the item was on the shelf two meters behind.
Leena usually buys a particular brand of soft drinks, which is now on promotion. The soft drink is not on her shopping list, but as she comes in the vicinity of the corresponding shelf, Ma$$iv€ tells her about the promotion. Using this kind of personalised and context-aware targeted advertisements, Leena avoids uninteresting advertisements and the promotions are timed to be displayed just at the time she is in close proximity to the product. When Leena is buying pasta, Ma$$iv€ tells which brand is cheapest per kg. She grabs one package and scans the barcode with the camera of her mobile device. A Wiki-type page opens, providing official information about the product, but also comments and recommendations from other users. She sees that the cheap pasta gets a bad recommendation from her Italian friend Carmelo. On her device, she selects the recommendation and chooses “I found the recommendation useful” from the feedback menu.
Leena’s shopping behaviour is learned on the basis of the information gathered in the preferred customer information system of the shopping chain. Such information is time-stamped, which has importance: Leena may buy different food for the weekend than for weekdays. Leena can also select between different types of recommendation profiles, for example: healthy, bargain, gourmet, season’s specials, or buddy-recommended
The main project research partners are:
- HIIT Adaptive Computing
- HIIT CoSCo
- HIIT Ubiquitous Interaction
- VTT Mobile Interaction Competence Center
The companies are:
- Bitlips Oy
- Ekahau Oy
- Elisa Oyj
- Idean Enterprises Oy
- Kesko Oyj
- Nokia Oyj
- Ramblas Digital Oy
- Tuulia International Oy
- Upcode Oy
- Finnish Federation of the Visually Impaired (Näkövammaisten keskusliitto, NKL)
Consortium Coordinator is Dr Patrik Floréen.