Classification of Soft Keyboard Typing Behaviors Using Mobile Device Sensors with Machine Learning

Yüksel, Asım Sinan; Şenel, Fatih Ahmet; Çankaya, İbrahim Arda

doi:10.1007/s13369-018-03703-8

Cited by 14 publications

(5 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Another study [60] used the same duration for the implementation of KD-based continuous authentication using EEG signals for desktop. Whereas Yuksel et al [390] used 1-minute duration to collect typing patterns via wallet apps for the same purpose. In case of predictive model, a study [226] used a 15-minute duration for identifying gender.…”

Section: G Time-span Of a Session In Continuous Modementioning

confidence: 99%

“…They have focused on incorporating additional features generated while playing a mobile game for Implicit Continuous Authentication. Another study [390], collected the sensor's data through a wallet app. They mainly focused on incorporating statistical features like minimum, maximum, mean, and standard deviation along with soft biometric features like age and gender.…”

Section: O Feature Fusion Approachesmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Literature Review on Latest Keystroke Dynamics Based Models

et al. 2022

View full text Add to dashboard Cite

The purpose of this study is to conduct a comprehensive evaluation and analysis of the most recent studies on the implications of keystroke dynamics (KD) patterns in user authentication, identification, and the determination of useful information. Another aim is to provide an extensive and up-to-date survey of the recent literature and potential research directions to understand the present state-of-the-art methodologies in this particular domain that are expected to be beneficial for the KD research community. From January 1st, 2017 to March 13th, 2022, the popular six electronic databases have been searched using a search criterion ("keystroke dynamics" OR "typing pattern") AND ("authentication" OR "verification" OR "identification"). With this criterion, a total of nine thousand three hundred forty-eight results, including duplicates, were produced. However, one thousand five hundred forty-seven articles have been chosen after removing duplicates and preliminary screening. Due to insufficient information, only one hundred twentyseven high-quality quantitative research articles have been included in the article selection process. We compared and summarised several factors with multiple tables to comprehend the various methodologies, experimental settings, and findings. In this study, we have identified six unique KD-based designs and presented the status of findings toward an effective solution in authentication, identification, and prediction. We have also discovered considerable heterogeneity across studies in each KD-based design for desktops and smartphones separately. Finally, this paper found a few open research challenges and provided some indications for a deeper understanding of the issues and further study.

show abstract

Section: G Time-span Of a Session In Continuous Modementioning

confidence: 99%

Section: O Feature Fusion Approachesmentioning

confidence: 99%

A Systematic Literature Review on Latest Keystroke Dynamics Based Models

et al. 2022

View full text Add to dashboard Cite

show abstract

“…k En yakın komşu (kNN), sınıflandırma sürecinde kullanılan denetimli bir sınıflandırma algoritmasıdır. Bu algoritma, yeni bir örneğin hangi sınıfa ait olduğuna, kullanıcı tarafından tanımlandığı kadar bir komşuya bakarak karar verme mantığıyla çalışmaktadır [31]. Bu çalışmada uzaklık değeri ölçülürken sıklıkla kullanılan Öklid uzaklık kriteri seçilmiştir.…”

Section: K En Yakın Komşuunclassified

Stress detection on smartphone data with a machine learning approach based on Mahalanobis distance-based outlier finding and ReliefF feature selection

Sağbaş¹,

Korukoğlu²,

Ballı³

2022

Pamukkale J Eng Sci

View full text Add to dashboard Cite

“…Ref. [9] achieved the highest sensitivity, specificity and accuracy of 86.95% in a bank dataset. Ref.…”

Section: Introductionmentioning

confidence: 97%

Improving the Accuracy of Predicting Bank Depositor’s Behavior Using a Decision Tree

Safarkhani

Moro

2021

Applied Sciences

View full text Add to dashboard Cite

Telemarketing is a widely adopted direct marketing technique in banks. Since customers hardly respond positively, data prediction models can help in selecting the most likely prospective customers. We aim to develop a classifier accuracy to predict which customer will subscribe to a long-term deposit proposed by a bank. Accordingly, this paper focuses on a combination of resampling, in order to reduce the imbalanced data, using feature selection, to reduce the complexity of data computing and dimension reduction of inefficiency data modeling. The performed operation has shown an improvement in the performance of the classification algorithm in terms of accuracy. The experimental results were run on a real bank dataset and the J48 decision tree achieved 94.39% accuracy prediction, with 0.975 sensitivity and 0.709 specificity, showing better results when compared to other approaches reported in the existing literature, such as logistic regression (91.79 accuracy; 0.975 sensitivity; 0.495 specificity) and Naive Bayes classifier (90.82% accuracy; 0.961 sensitivity; 0.507 specificity). Furthermore, our resampling and feature selection approach resulted in improved accuracy (94.39%) when compared to a state-of-the-art approach based on a fuzzy algorithm (92.89%).

show abstract

Classification of Soft Keyboard Typing Behaviors Using Mobile Device Sensors with Machine Learning

Cited by 14 publications

References 51 publications

A Systematic Literature Review on Latest Keystroke Dynamics Based Models

A Systematic Literature Review on Latest Keystroke Dynamics Based Models

Stress detection on smartphone data with a machine learning approach based on Mahalanobis distance-based outlier finding and ReliefF feature selection

Improving the Accuracy of Predicting Bank Depositor’s Behavior Using a Decision Tree

Contact Info

Product

Resources

About