Keystroke Patterns Research Articles

A Performance Benchmark for the PostgreSQL and MySQL Databases

This study highlights the necessity for efficient database management in continuous authentication systems, which rely on large-scale behavioral biometric data such as keystroke patterns. A benchmarking framework was developed to evaluate the PostgreSQL and MySQL databases, minimizing repetitive coding through configurable functions and variables. The methodology involved experiments assessing select and insert queries under primary and complex conditions, simulating real-world scenarios. Our quantified results show PostgreSQL’s superior performance in select operations. In primary tests, PostgreSQL’s execution time for 1 million records ranged from 0.6 ms to 0.8 ms, while MySQL’s ranged from 9 ms to 12 ms, indicating that PostgreSQL is about 13 times faster. For select queries with a where clause, PostgreSQL required 0.09 ms to 0.13 ms compared to MySQL’s 0.9 ms to 1 ms, making it roughly 9 times more efficient. Insert operations were similar, with PostgreSQL at 0.0007 ms to 0.0014 ms and MySQL at 0.0010 ms to 0.0030 ms. In complex experiments with simultaneous operations, PostgreSQL maintained stable performance (0.7 ms to 0.9 ms for select queries during inserts), while MySQL’s performance degraded significantly (7 ms to 13 ms). These findings underscore PostgreSQL’s suitability for environments requiring low data latency and robust concurrent processing capabilities, making it ideal for continuous authentication systems.

Extracting typing game keystroke patterns as potential indicators of programming aptitude

This study attempted to determine whether individuals possess programming aptitude solely based on keystroke information from typing games where participants type computer programs. The participants were students enrolled in university programming courses. The results indicated that using typing speed alone as an indicator achieved an accuracy of 0.71, while employing a custom machine learning model achieved an accuracy of 0.83. Additionally, it was found that individuals with programming aptitude tended to type the enter key relatively slower compared to other keys.

Efficacy of Keystroke Dynamics-Based User Authentication in the Face of Language Complexity

Introduction: This study investigates the impact of language complexity on Keystroke Dynamics (KD) and its implications for accurate KD-based user authentication system performance in smartphones. Method: This research meticulously analyzes keystroke patterns using 160 volunteers, including both frequently typed and infrequently typed texts. Our analysis of 12 anomaly detection algorithms reveals that a simple text-based KD system consistently outperforms its complex counterpart with superior Equal Error Rates (EERs). Results: As a result, the Scaled Manhattan anomaly detector achieves an EER of 2.48% for simple text and an improvement over 2.98% for complex text. The incorporation of soft biometrics further enhances algorithmic performance, emphasizing strategies to build resilience into KD-based user authentication systems. Conclusion: Throughout this study, the importance of text complexity is emphasized, and innovative pathways are introduced to strengthen KD-based user authentication paradigms.

Optimizing dynamic keystroke pattern recognition with hybrid deep learning technique and multiple soft biometric factors

In this work, we propose an optimization approach for dynamic keystroke pattern recognition by leveraging a hybrid deep learning technique and multiple soft biometric factors. Our methodology begins with the introduction of a novel algorithm called dynamic drone squadron optimization (DDSO) to optimize the selection of optimal features from a pool of multiple keystroke features. We then present an enhanced version of the improved sperm swarm optimization (ISSO) algorithm, which effectively combines the optimal weight features derived from multiple biometric responses. Furthermore, we introduce the multi-stage recurrent neural network (MS-RNN) classifier to accurately recognize and classify keystroke patterns. The performance of our proposed ISSO+MS-RNN technique is evaluated using the benchmark KBOC dataset to validate its effectiveness. Comparative analysis is conducted against existing state-of-the-art techniques, employing various evaluation measures, to demonstrate the superior performance of proposed approach.`

Identity authentication model from continuous keystroke pattern using CSO and LSTM network

Identity authentication model from continuous keystroke pattern using CSO and LSTM network

Identity authentication model from continuous keystroke pattern using CSO and LSTM network

Identity authentication model from continuous keystroke pattern using CSO and LSTM network

Revisiting the Security of Biometric Authentication Systems Against Statistical Attacks

The uniqueness of behavioral biometrics (e.g., voice or keystroke patterns) has been challenged by recent works. Statistical attacks have been proposed that infer general population statistics and target behavioral biometrics against a particular victim. We show that despite their success, these approaches require several attempts for successful attacks against different biometrics due to the different nature of overlap in users’ behavior for these biometrics. Furthermore, no mechanism has been proposed to date that detects statistical attacks. In this work, we propose a new hypervolumes-based statistical attack and show that unlike existing methods, it (1) is successful against a variety of biometrics, (2) is successful against more users, and (3) requires fewest attempts for successful attacks. More specifically, across five diverse biometrics, for the first attempt, on average our attack is 18 percentage points more successful than the second best (37% vs. 19%). Similarly, for the fifth attack attempt, on average our attack is 18 percentage points more successful than the second best (67% vs. 49%). We propose and evaluate a mechanism that can detect the more devastating statistical attacks. False rejects in biometric systems are common, and by distinguishing statistical attacks from false rejects, our defense improves usability and security. The evaluation of the proposed detection mechanism shows its ability to detect on average 94% of the tested statistical attacks with an average probability of 3% to detect false rejects as a statistical attack. Given the serious threat posed by statistical attacks to biometrics that are used today (e.g., voice), our work highlights the need for defending against these attacks.

Automatically detecting task-unrelated thoughts during conversations using keystroke analysis.

Task-unrelated thought (TUT), commonly referred to as mind wandering, is a mental state where a person’s attention moves away from the task-at-hand. This state is extremely common, yet not much is known about how to measure it, especially during dyadic interactions. We thus built a model to detect when a person experiences TUTs while talking to another person through a computer-mediated conversation, using their keystroke patterns. The best model was able to differentiate between task-unrelated thoughts and task-related thoughts with a kappa of 0.363, using features extracted from a 15 second window. We also present a feature analysis to provide additional insights into how various typing behaviors can be linked to our ongoing mental states.

KEYSTROKE DYNAMICS ANALYSIS USING MACHINE LEARNING METHODS

The primary objective of the paper was to determine the user based on its keystroke dynamics using the methods of machine learning. Such kind of a problem can be formulated as a classification task. To solve this task, four methods of supervised machine learning were employed, namely, logistic regression, support vector machines, random forest, and neural network. Each of three users typed the same word that had 7 symbols 600 times. The row of the dataset consists of 7 values that are the time period during which the particular key was pressed. The ground truth values are the user id. Before the application of machine learning classification methods, the features were transformed to z-score. The classification metrics were obtained for each applied method. The following parameters were determined: precision, recall, f1-score, support, prediction, and area under the receiver operating characteristic curve (AUC). The obtained AUC score was quite high. The lowest AUC score equal to 0.928 was achieved in the case of linear regression classifier. The highest AUC score was in the case of neural network classifier. The method of support vector machines and random forest showed slightly lower results as compared with neural network method. The same pattern is true for precision, recall and F1-score. Nevertheless, the obtained classification metrics are quite high in every case. Therefore, the methods of machine learning can be efficiently used to classify the user based on keystroke patterns. The most recommended method to solve such kind of a problem is neural network.

Detecting Cognitive Impairment Status Using Keystroke Patterns and Physical Activity Data among the Older Adults: A Machine Learning Approach.

Cognitive impairment has a significantly negative impact on global healthcare and the community. Holding a person's cognition and mental retention among older adults is improbable with aging. Early detection of cognitive impairment will decline the most significant impact of extended disease to permanent mental damage. This paper aims to develop a machine learning model to detect and differentiate cognitive impairment categories like severe, moderate, mild, and normal by analyzing neurophysical and physical data. Keystroke and smartwatch have been used to extract individuals' neurophysical and physical data, respectively. An advanced ensemble learning algorithm named Gradient Boosting Machine (GBM) is proposed to classify the cognitive severity level (absence, mild, moderate, and severe) based on the Standardised Mini-Mental State Examination (SMMSE) questionnaire scores. The statistical method “Pearson's correlation” and the wrapper feature selection technique have been used to analyze and select the best features. Then, we have conducted our proposed algorithm GBM on those features. And the result has shown an accuracy of more than 94%. This paper has added a new dimension to the state-of-the-art to predict cognitive impairment by implementing neurophysical data and physical data together.

Keystroke Dynamics Patterns While Writing Positive and Negative Opinions.

This paper deals with analysis of behavioural patterns in human–computer interaction. In the study, keystroke dynamics were analysed while participants were writing positive and negative opinions. A semi-experiment with 50 participants was performed. The participants were asked to recall the most negative and positive learning experiences (subject and teacher) and write an opinion about it. Keystroke dynamics were captured and over 50 diverse features were calculated and checked against the ability to differentiate positive and negative opinions. Moreover, classification of opinions was performed providing accuracy slightly above the random guess level. The second classification approach used self-report labels of pleasure and arousal and showed more accurate results. The study confirmed that it was possible to recognize positive and negative opinions from the keystroke patterns with accuracy above the random guess; however, combination with other modalities might produce more accurate results.

Dynamic keystroke pattern analysis and classifiers with competence for user recognition

The classification of biometric data is always a difficult task because these data are often unstable and depends on behavioural condition of a person. Personal keystroke activity is difficult to imitate and can be used for identity authentication. In this article, we suggest that the analysis of keystroke dynamics can dramatically increase the level of security of a computer system without disturbing the user’s comfort. First, the user’s established behavioural profile is loaded and then this profile is continuously compared with the current use of the keyboard. If a user’s temporary behavioural profile differs from their established profile, then the access to the computer is blocked. If a current profile of the user is recognized as legitimate, then profile can be dynamically changed, which allows to adjust profiles according to users abilities. Security based on keystroke dynamics analysis depends primarily on the quality of recognizing changes in a user’s profile. Recognition is performed on the dynamic analysis of the feature vector recorded while using the keyboard. In the presented approach, we propose using classifiers with their competence selection, which significantly improves the recognition of users. To select the most appropriate classifiers, the competences of the classifiers in a pool are calculated and then, the most powerful classifiers are selected. Conducted experiments, supported by the proper statistical analysis, confirm the usefulness of the proposed strategy in authorized user and intruder recognition.

EFFICIENCY OF BIOMETRIC RECOGNITION TECHNOLOGY BASED ON TYPING DYNAMICS IN MOOC

One of the problems that require a solution in Massive Open Online Courses (MOOC) is the lack of identification and authentication of the students. Different investigations have been carried out through several navigation, physiological and behavioral methods, achieving different recognition scales. Biometric authentication by keystroke patterns (Ups&Downs) has been implemented in several MOOCs for the ease of the digital platforms of the offeror to solve the identification problem. The objective of this research is to analyze the independence of the keystroke tool of the other demographic, sociographic and behavioral variables within a MOOC, establishing an initial pattern, and two authentication measurements throughout the course. The results show that the keystroke is independent of the analyzed variables, and it is reliable to identify the students in qualitative tests with extension answers.

S105. DIGITAL BIOMARKERS FOR THE ASSESSMENT OF COGNITIVE, BEHAVIORAL AND FUNCTIONAL OUTCOMES IN INDIVIDUALS WITH SCHIZOPHRENIA

BackgroundAn individual’s mental health is best captured by considering the overall associations of biological, behavioral, and social functions that comprise the framework of individual experience. As such, accessing data on various health indicators concurrently can influence prediction of disease progression or change in response to treatment. Data generated passively by smartphones to measure human behavior has generated significant research interest and has increasingly been utilized in psychiatric disorders. In schizophrenia, passive and continuous assessment of how an individual uses their mobile device may give rise to clinically useful markers that can be used to improve treatment processes, adapt treatment choices, identify early risk for relapse to initiate clinical intervention, and develop new clinical models. A promising approach is to leverage current advances in mobile technology, data analytics and machine learning to enable automated and fast phenotyping of digital data. . In this context, the workflow for phenotyping (passive data collection → data storage and curation → trait extraction → machine learning/classification → models/apps for decision support) has to be carefully designed and efficiently executed to minimize resource usage and maximize utility. Digital phenotyping can be used in conjunction with standard care to reduce time to recognition and acknowledgement that worsening of a symptom needs to be addressed, to reduce time to receiving appropriate level of care, to increase ability to analyze and collect data from a variety of sources to improve mental health needs assessment and delivery of services, and to advance outcome measurement through comparison of passive and active data sets.AimThe aim of this pilot study is to test whether a smartphone digital phenotyping application can help detect early signs of treatment failure or response in individuals with chronic schizophrenia after discharge from hospital.Methods17 individuals with DSM-5 schizophrenia were provided with a smartphone and digital phenotyping app, MindStrong Health app, following discharge from an inpatient psychiatric facility. The participants were followed for 6 months with monthly rater administered evaluations assessing neurocognition (Brief Assessment of Cognition in Schizophrenia (BACS)), symptomatology (PANSS; CGI-S/I), quality of life (SF-36), healthcare utilization, alcohol/drug use, level of clinical insight and depression (Calgary Depression Rating Scale, CDSS). Digital phenotyping data included gestures (swipes, taps, other), orientation (the way the phone is pointing), acceleration (sudden movements of the phone), keystroke patterns with characters encoded, number of calls made, number of emails sent, number of text messages sent, and location information from the GPS. Predictive models were built using multiple machine learning techniques - random forest plots, linear regression and gradient boosting, to predict the target scores based on phone usage patterns.ResultsOf the 17 enrolled participants, 10 provided analyzable data (i.e., had at least 22 target days with data). There was a gradual reduction of passive data generation due to either non-use of the smart phone or due to non-recharging of the device. The mean PANSS score was 80.12 (14.56). BACS scores corresponding to motor speed (token motor task), verbal fluency (category instances, letter fluency), and attention and processing speed (symbol coding) were found to be highly correlated with a composite digital phenotyping marker while scores on the PANSS or CDSS were not.DiscussionThe study provides a basis for using smartphone-base mobile apps to use as an augmentation within clinical practice to gather further information on patients outside the clinic setting, focusing on their behavior in the ‘real world.’ In particular, the cognitive data derived from the digital device correlated well with rater administered traditional cognitive ratings. Collecting digital data can provide a much needed window into the lives of patients in between normally scheduled visits, while minimizing costs and inconvenience to the patient. Studies with larger sample sizes are required to assess relationships with relapse, rehospitalizations and treatment failure.

Mood Detection from Physical and Neurophysical Data Using Deep Learning Models

Nowadays, smart devices as a part of daily life collect data about their users with the help of sensors placed on them. Sensor data are usually physical data but mobile applications collect more than physical data like device usage habits and personal interests. Collected data are usually classified as personal, but they contain valuable information about their users when it is analyzed and interpreted. One of the main purposes of personal data analysis is to make predictions about users. Collected data can be divided into two major categories: physical and behavioral data. Behavioral data are also named as neurophysical data. Physical and neurophysical parameters are collected as a part of this study. Physical data contains measurements of the users like heartbeats, sleep quality, energy, movement/mobility parameters. Neurophysical data contain keystroke patterns like typing speed and typing errors. Users’ emotional/mood statuses are also investigated by asking daily questions. Six questions are asked to the users daily in order to determine the mood of them. These questions are emotion‐attached questions, and depending on the answers, users’ emotional states are graded. Our aim is to show that there is a connection between users’ physical/neurophysical parameters and mood/emotional conditions. To prove our hypothesis, we collect and measure physical and neurophysical parameters of 15 users for 1 year. The novelty of this work to the literature is the usage of both combinations of physical and neurophysical parameters. Another novelty is that the emotion classification task is performed by both conventional machine learning algorithms and deep learning models. For this purpose, Feedforward Neural Network (FFNN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short‐Term Memory (LSTM) neural network are employed as deep learning methodologies. Multinomial Naïve Bayes (MNB), Support Vector Regression (SVR), Decision Tree (DT), Random Forest (RF), and Decision Integration Strategy (DIS) are evaluated as conventional machine learning algorithms. To the best of our knowledge, this is the very first attempt to analyze the neurophysical conditions of the users by evaluating deep learning models for mood analysis and enriching physical characteristics with neurophysical parameters. Experiment results demonstrate that the utilization of deep learning methodologies and the combination of both physical and neurophysical parameters enhances the classification success of the system to interpret the mood of the users. A wide range of comparative and extensive experiments shows that the proposed model exhibits noteworthy results compared to the state‐of‐art studies.

Efficient privacy-preserving implicit authentication

The number of online service accounts per person has rapidly increased over the last years. Currently, people have tens to hundreds of online accounts on average, and it is clear that users do not choose new, different, and strong passwords for each of these accounts. On the other hand, it is quite inconvenient for the user to be forced to explicitly authenticate each time she wants to use one of her many accounts; this is especially true with small user devices like smartphones. Implicit authentication is a way to mitigate the preceding problems by authenticating individuals based not only on their identity and credentials, but on how they interact with a device, i.e. their behavior. User behavior can be characterized by collecting keystroke patterns, browser history and configuration, IP addresses and location, among other characteristics of the user. However, keeping the user’s behavior profile in authentication servers can be viewed as privacy-invasive. Privacy-preserving implicit authentication has been recently introduced to protect the privacy of the users’ profiles, specifically against the party performing the authentication, which we call the server in the sequel. Yet, the privacy-preserving implicit authentication schemes proposed so far involve substantial computation both by the user and the server. We propose here a practical mechanism based on comparing behavior feature sets encoded as Bloom filters. The new mechanism entails much less computation and can accommodate much more comprehensive sets of features than previous alternatives.

Behavioral Biometrics for User Authentication Using Self-Refreshing SOM

Password protection has been a crucial issue in computer security. Traditional password protection using string does not provide sufficient security for user authentication. On the other hand, implementation of static biometrics in replace of password protection is costly as specific devices are needed. This research proposes the use of Self-Organizing Map (SOM) method for keystroke pattern recognition as an alternative to password protection. A self-refreshing SOM method was also implemented to increase the adaptability of SOM toward inconsistencies in typing pattern. Two main experiments were conducted to examine the ability of SOM for user recognition and the ability of self-refreshing SOM to deal with typing pattern inconsistencies. The first experiment shows the ability of SOM to authenticate users with very accurate classification. Users were successfully recognized and non-users were successfully rejected. Second experiment shows the ability of selfrefreshing SOM to deal with inconsistencies in typing pattern of a same user in which smoother transition of data trend was obtained.

CABA: Continuous Authentication Based on BioAura

Most computer systems authenticate users only once at the time of initial login, which can lead to security concerns. Continuous authentication has been explored as an approach for alleviating such concerns. Previous methods for continuous authentication primarily use biometrics, e.g., fingerprint and face recognition, or behaviometrics, e.g., key stroke patterns. We describe CABA, a novel continuous authentication system that is inspired by and leverages the emergence of sensors for pervasive and continuous health monitoring. CABA authenticates users based on their BioAura, an ensemble of biomedical signal streams that can be collected continuously and non-invasively using wearable medical devices. While each such signal may not be highly discriminative by itself, we demonstrate that a collection of such signals, along with robust machine learning, can provide high accuracy levels. We demonstrate the feasibility of CABA through analysis of traces from the MIMIC-II dataset. We propose various applications of CABA, and describe how it can be extended to user identification and adaptive access control authorization. Finally, we discuss possible attacks on the proposed scheme and suggest corresponding countermeasures.

English

With the fast development of video hardware and software in recent years, intelligent video systems have been widely used in industry, transportation, security, etc. at the same time, a lot of biometric technologies, comprising automated methods for uniquely recognizing people based on their physical or behavioural traits, such as face, fingerprint, palm print, finger-knuckle-print, gait, are also based on video or image analysis. Iris recognition is emerging as one of the important methods of biometrics-based identification systems. Several crucial factors Gait Keystroke pattern Signature of iris biometrics include rich and unique textures, non-invasiveness, and stability of iris pattern throughout the person’s lifetime, public acceptance, and availability of user friendly capturing devices. These factors have attracted the researchers to work in this evolving field over the past decade. The iris recognition consists of iris localization, normalization, encoding and comparison. In this paper, segmentation of periocular features and the encoding part of iris recognition is analysed.

An embedded system for extracting keystroke patterns using pressure sensors

Popular biometric security technologies include fingerprint and iris recognition systems. These technologies are extremely accurate because the patterns associated with an individual's finger or eye are very unique and static. However, when these technologies are used for physical access control they inform the potential adversary that specific characteristics are required to gain access. Behaviometrics aims to develop new strategies to enhance physical security via covert monitoring of distinct behavioral patterns. This research presents a novel stand-alone behaviometric prototype that incorporates standard password security with unique pressure characteristics to covertly analyse individual typing patterns. The prototype is evaluated under a controlled setting with 62 human subjects and nine classification algorithms. The kNN algorithm produced the highest classification rate of 94%. This research is one of the few papers that empirically substantiates biometric performance with a large-scale human subject trial, and also identifies several critical design considerations that impact classifier performance.