posted by organizer: portet || 12958 views || tracked by 5 users: [display]

SI : Multimodal Datasets in smart spaces 2021 : Special Issue: Multisensor and Multimodal Datasets in Intelligent Home for Context-Awareness, Human Home Interaction and Dialogue

FacebookTwitterLinkedInGoogle

Link: https://www.mdpi.com/journal/sensors/special_issues/multi_sensor_intelligent_home
 
When Nov 10, 2020 - Jun 1, 2021
Where Sensors journal
Submission Deadline Mar 31, 2021
Categories    pervasive computing   artificial intelligence   sensor networks   human computer interaction
 

Call For Papers

Intelligent home systems rely on sensor information and models of the underlying user behaviour to reason about the needs of the users and assist them in their everyday activities. With the continued shift from knowledge-based approaches to methods relying on machine learning techniques, the need of large-scale datasets of high quality increases. Datasets are essential both in designing and training the system to recognise and reason about the situation, either through the definition of a suitable situation model in knowledge-driven applications or through the preparation of training data for learning tasks in data-driven models. Hence, the quality of datasets can have a significant impact on the performance of the derived systems. Datasets are also vital for validating and quantifying the performance of applications. Furthermore, well documented datasets with reliable annotation ensure the reproducibility of research results. While datasets and the process of collecting and annotating them are highly developed in some fields, such as natural language processing (NLP) or computer vision, in smart homes, this data collection process is less systematic. This is due to several reasons:
- There is a large diversity of sensors and applications (highly variable input/output);
- The smart home domain is still a developing research field as opposed to some mature tasks such as in NLP;
- It is difficult to acquire ground truth in smart homes without jeopardizing the ecological aspect of the data being collected;
- There are no established roadmaps and standards for annotating data for smart homes, which results in datasets with different annotation granularity and structure, making them difficult to re-use in other applications/environments.

This call encourages submissions reporting methodological or experimental studies about ground truth labeled multisensor corpus collections. In particular, ecological corpus including several modalities (home automation sensors, wearable sensors, microphones, video cameras) and several dwellers are very welcome. All submissions should contain a section about the ethical positioning of the study/project. Further, when relevant, authors reporting a corpus collection are encouraged to make it available to the community and register it through a DOI. We also look forward to submissions which provide guidelines for sensor data collection and annotation or which empirically analyse existing or novel sensor datasets especially in the context of transfer learning.

The topics of interest include but are not limited to:
- Methods and intelligent tools for ecological corpus collection in intelligent or the general houses;
- Multimodal corpora (microphones, video cameras and any other sensors) made publicly available to the community;
- Processes of and best practices in collecting ecological user data;
- Improving and evaluating the quality of annotations;
- Ethical and privacy issues in data collection in home;
- Overview of currently available corpora in smart home;
- Synthetic data generation for improving models learning;
- Machine learning techniques to re-use mismatched data.


Important Dates
-----------------------
Submission Deadline: *March 31st, 2021*

Submission Guidelines
-------------------------------
Manuscripts should be submitted online at www.mdpi.com. Once you are registered, go to the submission form (https://susy.mdpi.com/user/manuscripts/upload/?journal=sensors). Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere. All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page (https://www.mdpi.com/journal/sensors/instructions). Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2000 CHF (Swiss Francs).

Guest Editors
-------------------
- Fran├žois Portet, Univ. Grenoble Alpes, France (francois.portet@imag.fr)
- Kristina Yordanova, University of Rostock, Germany (kristina.yordanova@text2hbm.org)
- Hirohiko Suwa, Nara Institute of Science and Technology (NAIST), Japan (h-suwa@po.wind.ne.jp)
- Emma Tonkin, University of Bristol, UK (e.l.tonkin@bristol.ac.uk)

For more details visit https://www.mdpi.com/journal/sensors/special_issues/multi_sensor_intelligent_home or contact one of the guest editors.

Related Resources

SI_DependXAIdata 2024   SI on Dependable and Reliable Datasets for Explainable Artificial Intelligence and Cyber Security Research
ECAI 2024   27th European Conference on Artificial Intelligence
ISEEIE 2024   2024 4th International Symposium on Electrical, Electronics and Information Engineering (ISEEIE 2024)
ACM-Ei/Scopus-CCISS 2024   2024 International Conference on Computing, Information Science and System (CCISS 2024)
Call For Papers Special Issue 2024   Smart Cities, innovating in the Transformation of Urban Environments
CCVPR 2024   2024 International Joint Conference on Computer Vision and Pattern Recognition (CCVPR 2024)
ML4CPS 2024   Machine Learning for Cyber-Physical Systems
SOFTFM 2024   3rd International Conference on Software Engineering Advances and Formal Methods
CSCT 2024   Congress on Smart Computing Technologies
COMIT 2024   8th International Conference on Computer Science and Information Technology