TY - GEN
T1 - Fuzzy Linguistic Summaries for Explaining Online Semi-Supervised Learning
AU - Kaczmarek-Majer, Katarzyna
AU - Casalino, Gabriella
AU - Castellano, Giovanna
AU - Leite, Daniel
AU - Hryniewicz, Olgierd
N1 - Funding Information:
Katarzyna Kaczmarek-Majer received funding from Small Grants Scheme (NOR/SGS/BIPOLAR/0239/2020-00) within the research project: “Bipolar disorder prediction with sensor-based semi-supervised Learning (BIPOLAR)”. Datasets considered in this paper were collected in the CHAD project “Smartphone-based diagnostics of phase changes in the course of bipolar disorder” (RPMA.01.02.00-14-5706/16-00) that was financed from EU funds (Regional Operational Program for Mazovia) in 2017-2018. The authors thank the researchers Karol Opara and Weronika Radziszewska from Systems Research Institute, Polish Academy of Sciences for their support in data preparation. Gabriella Casalino acknowledges funding from the European Union PON project Ricerca e Innovazione 2014-2020, DM 1062/2021.
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Intelligent systems for the medical domain often require processing data streams that evolve over time and are only partially labeled. At the same time, the need for explanations is of utmost importance not only due to various regulations, but also to increase trust among systems' users. In this work, an online data-driven learning method with focus on the explainability of evolving models equipped with incremental semi-supervised learning algorithms is considered. The proposed method combines: (i) the Dynamic Incremental Semi-Supervised Fuzzy C-Means (DISSFCM) algorithm to incrementally classify subsets of data; with (ii) Linguistic Summarization, which provides explanations of the classification results in terms of short sentences in a natural language. The approach has been illustrated for streaming data collected from voice calls of patients affected by Bipolar Disorder. The results show the effectiveness of the proposed method in classifying instances belonging to healthy and affective states, and explaining the approximate reasoning behind the classification of new acoustic data related to patients.
AB - Intelligent systems for the medical domain often require processing data streams that evolve over time and are only partially labeled. At the same time, the need for explanations is of utmost importance not only due to various regulations, but also to increase trust among systems' users. In this work, an online data-driven learning method with focus on the explainability of evolving models equipped with incremental semi-supervised learning algorithms is considered. The proposed method combines: (i) the Dynamic Incremental Semi-Supervised Fuzzy C-Means (DISSFCM) algorithm to incrementally classify subsets of data; with (ii) Linguistic Summarization, which provides explanations of the classification results in terms of short sentences in a natural language. The approach has been illustrated for streaming data collected from voice calls of patients affected by Bipolar Disorder. The results show the effectiveness of the proposed method in classifying instances belonging to healthy and affective states, and explaining the approximate reasoning behind the classification of new acoustic data related to patients.
KW - Acoustic Data
KW - Explainable Artificial Intelligence (XAI)
KW - Fuzzy Linguistic Summaries
KW - Fuzzy c-Means
KW - Linguistic Summarization
KW - Semi-supervised Online Learning
KW - Streaming data
UR - http://www.scopus.com/inward/record.url?scp=85147690944&partnerID=8YFLogxK
U2 - 10.1109/IS57118.2022.10019636
DO - 10.1109/IS57118.2022.10019636
M3 - Conference contribution
AN - SCOPUS:85147690944
T3 - 2022 IEEE 11th International Conference on Intelligent Systems, IS 2022
BT - 2022 IEEE 11th International Conference on Intelligent Systems, IS 2022
A2 - Atanassov, Krassimir T.
A2 - Doukovska, Lyubka
A2 - Kacprzyk, Janusz
A2 - Krawczak, Maciej
A2 - Owsinski, Jan W.
A2 - Sgurev, Vassil
A2 - Szmidt, Eulalia
A2 - Zadrozny, Slawomir
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th IEEE International Conference on Intelligent Systems, IS 2022
Y2 - 12 October 2022 through 14 October 2022
ER -