The Prior Art Database and Publishing service will be updated on Sunday, February 25th, from 1-3pm ET. You may experience brief service interruptions during that time.
Browse Prior Art Database

Method and System to Generate Small Typing-Errors-Prone Passwords Based on User Behaviour

IP.com Disclosure Number: IPCOM000247916D
Publication Date: 2016-Oct-11
Document File: 2 page(s) / 30K

Publishing Venue

The IP.com Prior Art Database


Method and system to generate small typing-errors-prone passwords based on user behaviour

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Method and System to Generate Small Typing -Errors-Prone Passwords Based on User Behaviour

    The article describes a new method to generate automatically user passwords. Nowadays the utmost focus on security increased the number of passwords that each person needs to handle in order to access a variety of devices and applications. In addition to this, the need of having "strong" password often means that these passwords are generated to contain a mix of numbers, capital and lower letters and alphanumeric symbols. No need to say that this increases the possibility of errors during typing. The objective of this article is to provide a method and system to generate passwords for whatever security application and profiling them in order to minimize the probability of typing errors for the specific user that will use the generated password.

    The core of the solution is based on known machine learning algorithm for detecting typing errors and the novel part that is the method and system to generate passwords with low probability of typing errors. Basically here is proposed a client application that uses machine learning to detect the most common typing errors done from a specific user. The outcome of this agent is an probability index that says how likely is that given a key on the keyboard, say "a", the user types "s" instead of "a" and this index is computed for each key on the keyboard. Based on this information, as detailed below an algorithm is provided which, given a set of security policies, generates a ranked list of passwords. The top ranked passwords are the ones for which the probability of typing errors for that uses is the smallest one. As outlined above the first part of the solution is based on a machine learning agent that does the following:

    1) When the agent is turned on, the agent detects every key pres...