Mathematical Model for Describing Familiarity
This visualization demonstrates a model developed for Intel’s Local Experiences of Automobility project, in which GPS and cell phone tracking data were combined with ethnographic research to build a rich and detailed picture of participants’ relationships with their vehicles. One of the major focus areas of the project was the question of routine: what makes a drive routine? The model demonstrated by this visualization provides a mathematical description of how familiar a user is with a particular area; in other words, it addresses the question of spatial routines.
For a given point, the model is calculated from two components: how close the point is to most of the points in the participant’s history, and whether the participant has ever visited that point before. By calculating this value for every point in the area a participant drove, we can generate heat maps like the one shown above. Lighter areas are areas with a higher familiarity score; darker areas have a lower familiarity score. The participant’s actual GPS paths are superimposed onto the heat map, so the relationship between the two can be seen. Using this model, we were able to investigate the relationship between a participant’s familiarity with an area and their phone usage.