Abstract
The Internet of Things (IoT) allows various embedded devices and smart sensors to be connected with each other, which provides a basis for building smart cities. The IoT-enabled smart city can greatly benefit people's daily lives, where smartphone is one of the most widely used IoT devices. For example, people can use the phone to check their financial account, store personal data and communicate with peers. Thus it is very important to safeguard the phones from unauthorized access. To complement traditional textual passwords, touch behavioral authentication has attracted much attention while it is still a challenge on how to build a robust scheme in practice. This is because users’ touch actions are often dynamic and hard to model. For this challenge, previous work has proved that touch actions could become consistent when users interact with social networking applications. Motivated by this observation, in this work, we perform a study to investigate users’ touch behavior within Email applications on smartphones (with Email being one of the most important and widely used means in connecting with others). The study results with 60 participants validate the former observation that users’ touch behavioral deviation can be greatly decreased when they play Email applications.
| Original language | English |
|---|---|
| Journal | Pattern Recognition Letters |
| Volume | 144 |
| Pages (from-to) | 35-41 |
| ISSN | 0167-8655 |
| DOIs | |
| Publication status | Published - Apr 2021 |
Bibliographical note
Funding Information:We would like to thank all participants for their hard work during the user study. This work was partially supported by National Natural Science Foundation of China (No. 61802077 ).
Publisher Copyright:
© 2021 Elsevier B.V.
Keywords
- Behavioral user authentication
- Machine learning
- Smartphone security
- Social networking
- Touch gestures
- Usable security
