Identifying Child Users via Touchscreen Interactions

Abstract

With the proliferation of smart devices, children can be easily exposed to violent or adult-only content on the Internet. Without any precaution, the premature and unsupervised use of smart devices can be harmful to both children and their parents. Thus, it is critical to employ parent patrol mechanisms such that children are restricted to child-friendly content only. A successful parent patrol strategy has to be user friendly and privacy aware. The apps that require explicit actions from parents are not effective because a parent may forget to enable them, and the ones that use built-in cameras or microphones to detect child users may impose privacy violations. In this article, we propose iCare, a system that can identify child users automatically and seamlessly when users operate smartphones. In particular, iCare investigates the intrinsic differences of screen-touch patterns between child and adult users from the aspect of physiological maturity. We discover that one’s touch behaviors are related to his or her age. Thus, iCare records the touch behaviors and extracts hand geometry, finger dexterity, and hand stability features that capture the age information. We conduct experiments on 100 people including 62 children (3 to 17 years old) and 38 adults (18 to 59 years old). Results show that iCare can achieve 96.6% accuracy for child identification using only a single swipe on the screen, and the accuracy becomes 98.3% with three consecutive swipes.

Fingerprinting and Personal Information Leakage from Touchscreen Interactions

“…Furthermore, we examined the performance of our models using a number of machine learning classifiers -Support Vector Machine, Random Forest, and Neural Network. The three classifiers were chosen as they are the best performing for touch-based authentication [16] and are commonly found in related work [4,10,14,30,34]:…”

“…Similarly, Jain et al [19] focused specifically on gender prediction improving on previous work and achieving ∼93% accuracy. Acien et al [2], Nguyen et al [30] and Cheng et al [10] used touch interaction data of children and adults to differentiate between the two groups. All three studies report accuracies of above 96%.…”

On the shape of timings distributions in free-text keystroke dynamics profiles

“…Most of the remaining studies consider the lognormal and other related distributions in a more general setting that includes not only keystroke dynamics but also touch-screen biometrics [27]. Going beyond authentication, [28] and [29] employ the sigma-lognormal model of rapid human movements to detect the age group of users based on their interaction with a touch screen, while [30] leverages different distributions to discriminate a human user from a bot. No other systematic comparison of distributions for the task of fitting keystroke timings histograms was found other than the aforementioned [21], [22], and [11 1.…”

Multi-Modal Deep Learning for Assessing Surgeon Technical Skill

“…Machine learning algorithms have been applied to surgical skills assessment by previous studies [ 4 , 7 , 9 , 12 , 16 ]. Classical machine learning, combining engineered features with learned classifiers, as well as deep learning models have shown promising results for both skills assessment works [ 17 ], as well as human activity recognition tasks (HAR) [ 18 , 19 , 20 , 21 , 22 , 23 ]. More recent work has focused on deep learning networks because of their ability to better exploit rich data sources (e.g., images, videos, motion tracking), which has led to improvements in performance.…”

“…Similarly to some image-based approaches [ 24 ], we employ a late-fusion approach. Some HAR studies investigate concatenating extracted features from different gestures for classical machine learning algorithms, and report that which features were extracted was more important than the fusion technique [ 23 ]. Unlike the studies in [ 23 , 24 ], we investigate fusing features extracted from disparate modalities (kinematic time-series + images) and not a single modality (images), and fuse learned features extracted from the raw data by the neural networks, instead of fusing hand-crafted features.…”