Page 130 - Kaleidoscope Academic Conference Proceedings 2022

2022 ITU Kaleidoscope Academic Conference

Smart mobile app

The mobile app is primarily designed for parents to receive all contextual and event information along with Machine Learning
(ML) based insights from cloud-based server. This information is mashed with the personal information of the concerned
child, securely entered in the app, to get insights about the child’s emotional wellness and identify any potential causes for
concern. Parents can choose the type and details of personal information to enter about their children. For example, the
parents can enter negative emotions the teenager is experiencing in the app and the information gets synthesized with XR
experiences to generate personalized interventions and recommendations.

Cloud-based server

A cloud-based server facilitates and synthesizes the information received from the mobile app and harvester to identify
potential problems and generate personalized recommendations. The data is encrypted and provided to authorized apps using
ephemeral tokens.

4. DEMO

The demo, which lasts for about 3 min, will be principally focused on the mobile app and how it provides actionable insights
for the parents. Also, the demo will highlight how the app could be used to customize for child’s sensitivities such as flashlight
sensitivities. Since the information captured and used in ParGuard is highly sensitive information, the demo will show how
differential privacy is enabled to protect the privacy of all parties involved.

REFERENCES
[1] https://www.govtech.com/policy/conn-bill-would-restrict-teenagers-social-media-use

[2] https://www.wsj.com/articles/tiktok-to-adjust-its-algorithm-to-avoid-negative-reinforcement-11639661801

[3] https://eprint.iacr.org/2020/340.pdf

[4] https://www.sciencedirect.com/science/article/pii/S0933365717306140

[5] Ishimaru S, Kunze K, Kise K, Weppner J, Dengel A, Lukowicz P, et al. In the Blink of an Eye: Combining Head
Motion and Eye Blink Frequency for Activity Recognition with Google Glass. In: ACM Augmented Human
International Conference. New York, NY, USA: ACM; 2014. p. 15:1–15:4.

– 84 –