User Understanding from Big Data Workshop (U2BigData)
Event submitted on Monday, October 12th 2020, approved by Henry Dalzel ✓
This event has been tagged as follows:
This would be an excellent conference for those with working with IOT technology and also living and working in Atlanta
The following description was either submitted by the Conference Organizer on Monday, October 12th 2020, or created by us.
The availability of massive amounts of data has driven significant progress in the field of AI, in particular, data-driven methods to understand human behavior has been an emerging topic in social science and human studies. Most internet companies need to leverage user-level data from different sources to understand how users interact with their products in various scenarios and contexts. Quantitative techniques would increase the generalizability of research conclusions regarding user mental models to provide frameworks for user understanding. On the other hand, large scale data can be critical to customize approaches in gaining user traction, improve user experience, and monetization for different user groups. We’ve seen tremendous applications in this space, including but not limited to the recommendation, marketing, online experiments, to name just a few. Fundamental understanding also requires methods such as statistical sampling, data visualization, funnel analysis, experimental design, causal inference, etc.
This workshop aims to provide a platform for researchers from related fields to exchange ideas on how to use data-driven technologies for better user understanding through data analytics & modeling, experimental design, and user research. The workshop will focus on both theoretical and practical challenges. Furthermore, it will place particular emphasis on algorithmic approaches in the context of learning, optimization, decision making, fairness, and data privacy that raise fundamental challenges for existing techniques.
Perspective and vision papers are also welcome. Finally, the workshop welcomes papers that describe the public release of privacy-preserving datasets that the community can use to solve fundamental technical problems of interest in user understanding.