Incremental Semi-Supervised Fuzzy C-Means for Bipolar Disorder Episode Prediction

Casalino, Gabriella; Dominiak, Monika; Galetta, Francesco; Kaczmarek-Majer, Katarzyna

doi:10.1109/eais48028.2020.9122748

Cited by 7 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the overall, the classification results achieved by DISSFCM on the two different sets of chunks are quite satisfying if we consider that BD episode prediction is a very difficult task due to several factors involved in the human brain activities. This is confirmed also by the empirical study in [5] where standard classification algorithms such as SVM, Decision Trees and Random Forest return relatively low accuracy values on the same dataset. The best accuracy achieved by Random Forest, with all the data labeled is of 67% on patient 1, and 57% on patient 2.…”

Section: Experimental Settingssupporting

confidence: 66%

“…In this work, we have investigated the effectiveness of the DISSFCM data stream classification method to predict Bipolar Disorder episodes on the basis of acoustic data collected during the interaction of patients with a dedicated smartphone application. Preliminary results on data of two different patients showed that DISSFCM provides good results with respect to the baseline provided by fully supervised learning methods applied to the same BD data [5], even when working on partially labeled data. Compared to its previous static versions, DISSFCM shows a better capability in capturing changes in data thanks to a splitting mechanism that adapts the number of clusters.…”

Section: Discussionmentioning

confidence: 99%

“…Alternatively to state-of-art methods that use supervised or unsupervised learning approaches, in [5] we initially explored the use of a semi-supervised learning approach that combines both labeled and unlabeled data to derive a model for BD episode prediction. Specifically, the ISSFCM (Incremental Semi-Supervised Fuzzy C-Means) [6] and its batch version SSFCM [30] were compared with standard classification algorithms to predict bipolar disorder.…”

Section: Introductionmentioning

confidence: 99%

“…ISSFCM offers the possibility to process in an incremental fashion the data stream of the voice signal obtained from the interaction of the patients with their smartphone, thus exploiting the time structure of data which is ignored by static learning methods [18]. Moreover, the results in [5] showed that ISSFCM achieves good results thanks to fuzzy clustering that enhances the detection of patterns in data and provides interpretable results in terms of labeled prototypes that represent data in a synthetic manner.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Dynamic Incremental Semi-supervised Fuzzy Clustering for Bipolar Disorder Episode Prediction

et al. 2020

Self Cite

View full text Add to dashboard Cite

Bipolar Disorder (BD) is a chronic mental illness characterized by changing episodes from euthymia (healthy state) through depression and mania to the mixed states. In this context, data collected through the interaction of patients with smartphones enable the creation of predictive models to support the early prediction of a starting episode. Previous research on predicting a new BD episode use mostly supervised learning methods that require labeled data and hence force a filtering of the available data to retain only those data that have valid labels (from the psychiatric assessment). To avoid limitations of supervised learning, in this paper we investigate the use of a semi-supervised learning approach that combines both labeled and unlabeled data to derive a model for BD episode prediction. Specifically we apply the DISSFCM (Dynamic Incremental Semi-Supervised Fuzzy C-Means) algorithm which offers the possibility to process in an incremental fashion the data stream of the voice signal captured by the smartphone, thus exploiting the evolving time structure of data which is ignored by static learning methods. DISS-FCM processes data in form of chunks and creates a dynamic collection of clusters thanks to a splitting mechanism that generates new clusters to better capture the hidden geometrical structure of data. This gives DISSFCM the ability to detect changes in data and dynamically adapt the model to them, thus improving the prediction accuracy. Preliminary results on real-world data collected at the Department of Affective Disorders, Institute of Psychiatry and Neurology in Warsaw (Poland) show that DISSFCM is able to predict some of healthy episodes (euthymia) and disease episodes even when only 25% of labeled data are available. Moreover DISSFM performs better than its previous version without split (ISSFCM) and it also overcomes the batch algorithm (SSFCM) that uses the whole dataset to create the model.

show abstract

Section: Experimental Settingssupporting

confidence: 66%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Dynamic Incremental Semi-supervised Fuzzy Clustering for Bipolar Disorder Episode Prediction

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Works [5]- [13] extended SSFCMeans in different ways and modified the mechanism of handling partial supervision to various extents, but did not change the core idea of additive combination nor its interpretation. Works [14]- [17] wrapped SSFCMeans to analyze data streams, primarily in the problem of monitoring bipolar disorder. Works [18]- [22] explored safe semi-supervised clustering aiming at handling mislabeled instances (label errors).…”

Section: Introductionmentioning

confidence: 99%

Explainable Impact of Partial Supervision in Semi-Supervised Fuzzy Clustering

Kmita,

Kaczmarek-Majer,

Hryniewicz

2024

IEEE Trans. Fuzzy Syst.

Self Cite

View full text Add to dashboard Cite

Controlling the impact of partial supervision on the outcomes of modeling is of uttermost importance in semisupervised fuzzy clustering. Semi-Supervised Fuzzy C-Means (SSFCMeans), a specific model we consider, uses a single hyperparameter called a scaling factor α to weigh the impact of partially labeled data. This concept became widespread and was reused directly in many works building on SSFCMeans, or even applied to other fuzzy clustering algorithms such as Possibilistic C-Means. However, none of the works challenged the original interpretation of α which suggests that the impact of partial supervision is directly proportional to the scaling factor. We fill the above research gap and thoroughly analyze this relationship. We provide novel explanations of the scaling factor α in terms of the key element of fuzzy clustering -the membership values. We prove that the impact of partial supervision is a non-linear function of α. Our approach is rooted in the explainability framework, which distinguishes interpretation from an explanation and treats the latter as superior. Explaining the scaling factor leads to an explainable impact of partial supervision and enables greater control of it. Finally, built on the novel explanations, we propose a unified, analytically justified framework for selecting the value of the hyperparameter α that is based on the crossvalidation approach. We illustrate that the proposed framework enables an extensive analysis of the impact of partial supervision in SSFCMeans with a simulation experiment.

show abstract