University of Birmingham > Talks@bham > Artificial Intelligence and Natural Computation seminars > Population-scale Mobile Sensing Systems

Population-scale Mobile Sensing Systems

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Lars Kunze.

Host: Mirco Musolesi, PhD --- Please note: This is a joint AINC/HCI seminar.

Through the widespread adoption of sensor-enabled smartphones and wearables, it is now feasible to build crowd-based systems able to gather and reason over large-scale mobile sensor data. These systems have the potential to transform many existing consumer services and data-driven organizations such as city planning, public health and environmental agencies. However, a variety of technical and social barriers must be overcome for this potential to be realized. For example, sensing systems often require sustained user attention and engagement to collect or interpret data. Furthermore, they must cope with extreme levels of user (e.g., demographics, lifestyles) and context diversity when deployed at large-scale.

To address these barriers, in this talk I will present two key ideas. The first idea – CrowdSense@Place (CSP) – combines data collection, transparent to the user, and limited crowd labeling with a series of multi-modal classifiers towards a challenging representative sensing problem: linking place visits to place categories (e.g., shopping, gym, and restaurant). The second idea – Community Similarity Networks (CSN) – is an activity recognition framework that exploits crowdsourced sensor data to personalize classifiers with data contributed from other similar users. CSN incorporates inter-person similarity measurements into the classifier training process resulting in increased robustness to complex environments and heterogeneous user populations. Together techniques like CSP and CSN seek to push mobile sensing forward – enabling it to potentially scale beyond 100s of users to potentially 100s of millions.

This talk is part of the Artificial Intelligence and Natural Computation seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

Talks@bham, University of Birmingham. Contact Us | Help and Documentation | Privacy and Publicity.
talks@bham is based on talks.cam from the University of Cambridge.