Identification of programmers from typing patterns

Abstract

Being able to identify the user of a computer solely based on their typing patterns can lead to improvements in plagiarism detection, provide new opportunities for authentication, and enable novel guidance methods in tutoring systems. However, at the same time, if such identification is possible, new privacy and ethical concerns arise. In our work, we explore methods for identifying individuals from typing data captured by a programming environment as these individuals are learning to program. We compare the identification accuracy of automatically generated user profiles, ranging from the average amount of time that a user needs between keystrokes to the amount of time that it takes for the user to press specific pairs of keys, digraphs. We also explore the effect of data quantity and different acceptance thresholds on the identification accuracy, and analyze how the accuracy changes when identifying individuals across courses. Our results show that, while the identification accuracy varies depending on data quantity and the method, identification of users based on their programming data is possible. These results indicate that there is potential in using this method, for example, in identification of students taking exams, and that such data has privacy concerns that should be addressed.

Educational Data Mining and Learning Analytics in Programming

“…For example, source code files can include identifying information in comments, and the identifiers in the program code (student invented names of variables and data types) can contain information that may destroy the anonymity of the data set. Moreover, even the time between key presses can be used to distinguish between students [76]. All these aspects should be taken care of when data sets are prepared for publicity.…”

Autenticação contínua de alunos utilizando biometria comportamental em ambiente Juiz On-line

“…Não foram encontrados trabalhos que abordem a utilização da dinâmica codificação para a autenticação contínua de alunos em ambientes de juiz on-line. Entretanto, o estudo de [Longi et al 2015] mostra que os alunos podem ser autenticados ou identificados através destas características em atividades das disciplinas de programação. Nesse estudo, os autores estudaram o quanto a quantidade dos dados afeta a precisão da identificação dos alunos.…”

Using Keystroke Analytics to Improve Pass-Fail Classifiers

“…Indeed, they note that while many better coders type fast, some poor coders also exhibited a rapid keystroke frequency. Longi et al (2015) also use keystroke metrics so solve a slightly different problem, that of identifying a programmer from their keystroke timings. The authors used a number of approaches, with the most complex being to build a profile of the digraph timings for each programmer.…”

“…For example, during coding sessions, students are encouraged to interrupt their typing to sketch out ideas or to consult notes. Similar to the approach taken by Dowland and Furnell (2004) and Longi et al (2015), a data pre-processing stage was applied to address these outliers. An upper bound of 2 seconds on the digraphs was applied, eliminating all digraphs with latencies greater than this.…”