Dynamic Active Learning based on agreement and applied to emotion recognition in spoken interactions. Zhang, Y., Coutinho, E., Zhang, Z., Quan, C., & Schuller, B. In ICMI 2015 - Proceedings of the 2015 ACM International Conference on Multimodal Interaction, pages 275-278, 2015. ACM Press.
Dynamic Active Learning based on agreement and applied to emotion recognition in spoken interactions [link]Website  abstract   bibtex   
In this contribution, we propose a novel method for Active Learning (AL) - Dynamic Active Learning (DAL) - which targets the reduction of the costly human labelling work necessary for modelling subjective tasks such as emotion recognition in spoken interactions. The method implements an adaptive query strategy that minimises the amount of human labelling work by deciding for each instance whether it should automatically be labelled by machine or manually by human, as well as how many human annotators are required. Extensive experiments on standardised test-beds show that DAL significantly improves the efficiency of conventional AL. In particular, DAL achieves the same classification accuracy obtained with AL with up to 79.17 % less human annotation effort.
@inproceedings{
 title = {Dynamic Active Learning based on agreement and applied to emotion recognition in spoken interactions},
 type = {inproceedings},
 year = {2015},
 identifiers = {[object Object]},
 keywords = {article,conference},
 pages = {275-278},
 websites = {http://dl.acm.org/citation.cfm?doid=2818346.2820774},
 publisher = {ACM Press},
 city = {New York, New York, USA},
 id = {b81a6392-0597-30a6-a8ba-10abc421d190},
 created = {2020-05-29T11:51:38.588Z},
 file_attached = {true},
 profile_id = {ffa9027c-806a-3827-93a1-02c42eb146a1},
 last_modified = {2020-05-30T17:16:24.065Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {zhang2015dynamicinteractions},
 source_type = {inproceedings},
 folder_uuids = {aac08d0d-38e7-4f4e-a381-5271c5c099ce},
 private_publication = {false},
 abstract = {In this contribution, we propose a novel method for Active Learning (AL) - Dynamic Active Learning (DAL) - which targets the reduction of the costly human labelling work necessary for modelling subjective tasks such as emotion recognition in spoken interactions. The method implements an adaptive query strategy that minimises the amount of human labelling work by deciding for each instance whether it should automatically be labelled by machine or manually by human, as well as how many human annotators are required. Extensive experiments on standardised test-beds show that DAL significantly improves the efficiency of conventional AL. In particular, DAL achieves the same classification accuracy obtained with AL with up to 79.17 % less human annotation effort.},
 bibtype = {inproceedings},
 author = {Zhang, Yue and Coutinho, Eduardo and Zhang, Zixing and Quan, Caijiao and Schuller, Björn},
 booktitle = {ICMI 2015 - Proceedings of the 2015 ACM International Conference on Multimodal Interaction}
}

Downloads: 0