Non-Parametric Clustering Using Deep Neural Networks

Avgerinos, Christos; Solachidis, V.; Vretos, Nicholas; Daras, Petros

doi:10.1109/access.2020.3017884

Cited by 4 publications

(2 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Except for Scenario1 and Scenario2 where the number of clusters is given, some authors assume that the number of clusters is not given (Chen, 2015;Yang et al, 2016;Caron et al, 2018;Mautz et al, 2019;Avgerinos et al, 2020). For example, in DLNC (Chen, 2015), for a given unlabeled dataset, the feature is extracted by a deep belief network.…”

Section: Appendix a Review Of Related Workmentioning

confidence: 99%

Deep Clustering with a Constraint for Topological Invariance based on Symmetric InfoNCE

Zhang¹,

Wada²,

Waida³

et al. 2023

Preprint

View full text Add to dashboard Cite

We consider the scenario of deep clustering, in which the available prior knowledge is limited. In this scenario, few existing state-of-the-art deep clustering methods can perform well for both non-complex topology and complex topology datasets. To address the problem, we propose a constraint utilizing symmetric InfoNCE, which helps an objective of deep clustering method in the scenario train the model so as to be efficient for not only non-complex topology but also complex topology datasets. Additionally, we provide several theoretical explanations of the reason why the constraint can enhances performance of deep clustering methods. To confirm the effectiveness of the proposed constraint, we introduce a deep clustering method named MIST, which is a combination of an existing deep clustering method and our constraint. Our numerical experiments via MIST demonstrate that the constraint is effective. In addition, MIST outperforms other state-of-the-art deep clustering methods for most of the commonly used ten benchmark datasets.

show abstract

Section: Appendix a Review Of Related Workmentioning

confidence: 99%

Deep Clustering with a Constraint for Topological Invariance based on Symmetric InfoNCE

Zhang¹,

Wada²,

Waida³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…They also concluded that another aspect of willingness to use wearables was easy access and better connectivity to the healthcare system but did not explore this further. The TeNDER system directly addresses these points, as it facilitates seamless interaction between patients, caregivers, and healthcare professionals, fostering continuous awareness and real-time updates on patients' health status [12,13].…”

Section: Introductionmentioning

confidence: 99%

Personalizing Digital Health: Adapting Health Technology Systems to Meet the Needs of Different Older Populations

Jimenez¹,

Río

Berman³

et al. 2023

Healthcare

View full text Add to dashboard Cite

The ageing of the population is growing significantly and will challenge healthcare systems. Chronic diseases in the older population require a change in service delivery, and new technologies can be a key element in ensuring the viability and sustainability of these systems. However, the generation gap and the physical and cognitive decline commonly associated with the older generation are barriers to the transition to these models of care. Despite this, there has been a trend towards digital healthcare, which has many potential benefits for the older population. Numerous studies have assessed the acceptability of new technologies for older people in healthcare. These studies highlight the importance of perceived usefulness, compatibility, ease of use and personalisation of the technology. Personalisation is necessary to ensure that the system is useful for users, and different characteristics such as country of origin, gender, age, or comfort with the technology should be taken into account. A person-centred approach in the development of new health technology systems is essential to ensure that applications can be better tailored to the needs of different ageing populations. Many organisations have dedicated time and resources to ensure a person-centred approach in the development of new health technology systems, and putting the individual first is the best way forward in digital health. This article presents the work carried out in this regard in the framework of the European TeNDER project together with an analysis of the results obtained in terms of satisfaction, usefulness, and usability from end-users. The dynamic and continuous process carried out throughout the TeNDER project translates the needs reported by users, as far as personalisation of interactions is concerned. All end-users held a positive opinion about the usability and usefulness of the system.

show abstract