Seminar Human-Building Interaction

Supervisors: Denis Lalanne, Julien Nembrini

Big data buildings

Human Building Interaction aims at designing interactive building systems and the corresponding human-machine interfaces, and at evaluating them with inhabitants. The major goals of this multidisciplinary domain (informatics, psychology, sociology, design) are to improve the quality of life of residents, reduce energy consumption, and increase social and energy awareness.

Currently built buildings are more and more fitted with sensors and automation. These systems produce large amounts of data, which tends to be underexploited. Taking a data-centric point of view, the seminar will concentrate on questions related to acquisition, processing and final purpose of data in context of understanding building occupants' behaviour.

We invite 4 students in informatics to explore the current state of the art in research about big data in the built environment. Each student will be asked to choose a theme within the domain, select state-of-the-art references relevant to the chosen thematic, and synthesize these references in a written report, an oral presentation, as well as combining them in a quality article written collaboratively by the seminar participants.

Proposed sub-topics for big data buildings

  • General data acquisition techniques
  • Acquisition of occupants' behaviour
  • Data algorithms and analysis
  • Big data purpose & applications

Reference articles

The link to download the reference articles will be communicated in class.

Structure

The seminar will be divided into 2 parts:

  • Individual writing
  • Collaborative writing

In the first part, participants will be asked to :

  1. select a subtopic.
  2. read the proposed papers.
  3. select additional papers relevant to the subtopic by exploring papers referenced in the proposed ones (downstream) and papers that 4. reference the proposed papers (upstream).
  4. write a short paper (2 pages) on their subtopic authored in LaTeX following ACM Strict format.
  5. review the paper of a colleague providing constructive comments.
  6. present both short paper and review orally
  7. revise their paper according to the review from colleagues

In the second part, participants will be assigned an extended paragraph to write on their subtopic to write a scientific article collaboratively with the teaching staff. Provided sufficient quality is reached at the end of the semester, this article will be submitted to a conference in the research field, with all students and teaching staff as co-authors.

Learning Outcomes

At the end of the seminar, students will know how to do a bibliographic research and how to write a scientific article. They will also be exposed to the process of collaborative writing, as is standard practice in the research community. Further, they will know what is "Human Building Interaction" and the current techniques and trends, and will deepen their knowledge on the particular subtopic of big data.

Schedule

The seminar is on appointment. The schedule will be defined at the first seminar session. About 6 meetings will take place including presentations.

Because of the participative approach, the seminar is limited to 4 participants. We will therefore select the most motivated students. Please contact us soon with 3-4 lines explaining your motivation to participate to this seminar: julien [d-o-t] nembrini [A-T] unifr [d-o-t] ch

First seminar session: Wednesday September 20th 2017

Date: 2017, Fall