Balancing Activity Recognition and Privacy Preservation with a Multi-objective Evolutionary Algorithm

Poli, Angelica; Muñoz-Antón, Ángela M.; Spinsante, Susanna; Flórez-Revuelta, Francisco

doi:10.1007/978-3-030-91421-9_1

Cited by 3 publications

(1 citation statement)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A quite recent approach, not yet so much widespread in the market, is based on the Integrated Circuit Metric (ICMetric) technology that exploits features of a device to generate an identification, which is then used for the provision of cryptographic services (Tahir et al 2018 ). An alternative way to address the problem of protecting the user’s privacy, which is being investigated in the framework of the PAAL project, focuses on identifying those features inherently present within the collected signals, like in the case of the acceleration measured from a wrist-worn device, with the aim of removing those features, which may expose personal details about the person, without hindering the performance of automatic ADL classification algorithms (Poli et al 2021 ).…”

Section: Realization Of Pbd Approach Using Exemplary Lifelogging Appl...mentioning

confidence: 99%

Interdisciplinary perspectives on privacy awareness in lifelogging technology development

Wilkowska

Heek

Colonna

et al. 2022

J Ambient Intell Human Comput

Self Cite

View full text Add to dashboard Cite

Population aging resulting from demographic changes requires some challenging decisions and necessary steps to be taken by different stakeholders to manage current and future demand for assistance and support. The consequences of population aging can be mitigated to some extent by assisting technologies that can support the autonomous living of older individuals and persons in need of care in their private environments as long as possible. A variety of technical solutions are already available on the market, but privacy protection is a serious, often neglected, issue when using such (assisting) technology. Thus, privacy needs to be thoroughly taken under consideration in this context. In a three-year project PAAL (‘Privacy-Aware and Acceptable Lifelogging Services for Older and Frail People’), researchers from different disciplines, such as law, rehabilitation, human-computer interaction, and computer science, investigated the phenomenon of privacy when using assistive lifelogging technologies. In concrete terms, the concept of Privacy by Design was realized using two exemplary lifelogging applications in private and professional environments. A user-centered empirical approach was applied to the lifelogging technologies, investigating the perceptions and attitudes of (older) users with different health-related and biographical profiles. The knowledge gained through the interdisciplinary collaboration can improve the implementation and optimization of assistive applications. In this paper, partners of the PAAL project present insights gained from their cross-national, interdisciplinary work regarding privacy-aware and acceptable lifelogging technologies.

show abstract

Section: Realization Of Pbd Approach Using Exemplary Lifelogging Appl...mentioning

confidence: 99%

Interdisciplinary perspectives on privacy awareness in lifelogging technology development

Wilkowska

Heek

Colonna

et al. 2022

J Ambient Intell Human Comput

Self Cite

View full text Add to dashboard Cite

show abstract

Complex Human Activity Recognition Using a Local Weighted Approach

Asuroglu

2022

IEEE Access

View full text Add to dashboard Cite

Due to the life expectancy increase, there will be a workforce shortage in elderly care sector in forthcoming years. Ambient Assisted Living (AAL) systems can cope with this issue. A subset of AAL, Human activity recognition (HAR) provides an efficient way to tackle this issue. It can help with evaluating general health and welfare of elderly by automatically tracking their activities. Lifelogging and home diary applications will reduce the load on physicians and caregivers. On the other hand, complex activities play a vital role as they have high level semantic characteristics that truly represent daily life of the user. The main objective is to track these high-level semantic motions with low-cost single sensor systems with efficient machine learning frameworks. To achieve this objective, a framework is proposed to predict complex human activities from a single sensor using a machine learning approach. Time and frequency features are extracted from PAAL ADL Accelerometry Dataset and fed to Locally Weighted Random Forest (LWRF) machine learning algorithm. This algorithm is a hybrid structure that utilizes local weighting by introducing neighboring samples on Random Forest tree building phases. Proposed approach achieved 91% accuracy for HAR and 91.3% for gender recognition, outperforming other machine learning algorithms and previous study on the same dataset. This is the first study that utilize a local weighted approach for accelerometer signal domain. For prospective application, proposed framework can be embedded in lifelogging and home diary applications in home environments to track mental status of elderlies.

show abstract

Privacy-Preserving Human Action Recognition with a Many-Objective Evolutionary Algorithm

Climent-Pérez

Flórez-Revuelta

2022

Sensors

View full text Add to dashboard Cite

Wrist-worn devices equipped with accelerometers constitute a non-intrusive way to achieve active and assisted living (AAL) goals, such as automatic journaling for self-reflection, i.e., lifelogging, as well as to provide other services, such as general health and wellbeing monitoring, personal autonomy assessment, among others. Human action recognition (HAR), and in particular, the recognition of activities of daily living (ADLs), can be used for these types of assessment or journaling. In this paper, a many-objective evolutionary algorithm (MaOEA) is used in order to maximise action recognition from individuals while concealing (minimising recognition of) gender and age. To validate the proposed method, the PAAL accelerometer signal ADL dataset (v2.0) is used, which includes data from 52 participants (26 men and 26 women) and 24 activity class labels. The results show a drop in gender and age recognition to 58% (from 89%, a 31% drop), and to 39% (from 83%, a 44% drop), respectively; while action recognition stays closer to the initial value of 68% (from: 87%, i.e., 19% down).

show abstract

Balancing Activity Recognition and Privacy Preservation with a Multi-objective Evolutionary Algorithm

Cited by 3 publications

References 42 publications

Interdisciplinary perspectives on privacy awareness in lifelogging technology development

Interdisciplinary perspectives on privacy awareness in lifelogging technology development

Complex Human Activity Recognition Using a Local Weighted Approach

Privacy-Preserving Human Action Recognition with a Many-Objective Evolutionary Algorithm

Contact Info

Product

Resources

About