2012 | OriginalPaper | Buchkapitel
A Study on Noisy Typing Stream Analysis Using Machine Learning Approach
verfasst von : Jun Li
Erschienen in: Enterprise Information Systems
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
People’s behaviors on using computer keyboard are different. This is particularly the case within disabled community. The differences are reflected by individual’s distinct typing characters such as speed and error patterns, and the environment around. This paper studies features such as keyboard layout, key distance and time gap and provides evidence that these features significantly affect people’s typing performance. A specific user typing behavior, i.e. ‘Hitting Adjacent Key Errors’, is selected from categorized typing behaviors and simulated based on a probabilistic neural network algorithm to correct typing mistakes. Results demonstrate a high performance of the designed model, about 70% of all tests score above Basic Correction Rate, and simulation also shows a very unstable trend of user’s ‘Hitting Adjacent Key Errors’ behavior with specific datasets used by the research. Further work is suggested in the conclusion.