Smart Chatbot for User Authentication
Despite being the most widely used authentication mechanism, password-based authentication is not very secure, being easily guessed or brute-forced. To address this, many systems which especially value security adopt Multi-Factor Authentication (MFA), in which multiple different authentication mecha...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/23/4016 |
| _version_ | 1797463336012480512 |
|---|---|
| author | Peter Voege Iman I. M. Abu Sulayman Abdelkader Ouda |
| author_facet | Peter Voege Iman I. M. Abu Sulayman Abdelkader Ouda |
| author_sort | Peter Voege |
| collection | DOAJ |
| description | Despite being the most widely used authentication mechanism, password-based authentication is not very secure, being easily guessed or brute-forced. To address this, many systems which especially value security adopt Multi-Factor Authentication (MFA), in which multiple different authentication mechanisms are used concurrently. JitHDA (Just-in-time human dynamics based authentication engine) is a new authentication mechanism which can add another option to MFA capabilities. JitHDA observes human behaviour and human dynamics to gather up to date information on the user from which authentication questions can be dynamically generated. This paper proposes a system that implements JitHDA, which we call Autonomous Inquiry-based Authentication Chatbot (AIAC). AIAC uses anomalous events gathered from a user’s recent activity to create personalized questions for the user to answer, and is designed to improve its own capabilities over time using neural networks trained on data gathered during authentication sessions. Due to using the user’s recent activity, they will be easy for the authentic user to answer and hard for a fraudulent user to guess, and as the user’s recent history updates between authentication sessions new questions will be dynamically generated to replace old ones. We intend to show in this paper that AIAC is a viable implementation of JitHDA. |
| first_indexed | 2024-03-09T17:49:12Z |
| format | Article |
| id | doaj.art-390e2476700642a7ab602b88815a8ecb |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-09T17:49:12Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-390e2476700642a7ab602b88815a8ecb2023-11-24T10:49:17ZengMDPI AGElectronics2079-92922022-12-011123401610.3390/electronics11234016Smart Chatbot for User AuthenticationPeter Voege0Iman I. M. Abu Sulayman1Abdelkader Ouda2Department of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, CanadaDepartment of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, CanadaDepartment of Electrical and Computer Engineering, Western University, London, ON N6A 5B9, CanadaDespite being the most widely used authentication mechanism, password-based authentication is not very secure, being easily guessed or brute-forced. To address this, many systems which especially value security adopt Multi-Factor Authentication (MFA), in which multiple different authentication mechanisms are used concurrently. JitHDA (Just-in-time human dynamics based authentication engine) is a new authentication mechanism which can add another option to MFA capabilities. JitHDA observes human behaviour and human dynamics to gather up to date information on the user from which authentication questions can be dynamically generated. This paper proposes a system that implements JitHDA, which we call Autonomous Inquiry-based Authentication Chatbot (AIAC). AIAC uses anomalous events gathered from a user’s recent activity to create personalized questions for the user to answer, and is designed to improve its own capabilities over time using neural networks trained on data gathered during authentication sessions. Due to using the user’s recent activity, they will be easy for the authentic user to answer and hard for a fraudulent user to guess, and as the user’s recent history updates between authentication sessions new questions will be dynamically generated to replace old ones. We intend to show in this paper that AIAC is a viable implementation of JitHDA.https://www.mdpi.com/2079-9292/11/23/4016machine learningauthenticationnatural language understandingbig datachatbots |
| spellingShingle | Peter Voege Iman I. M. Abu Sulayman Abdelkader Ouda Smart Chatbot for User Authentication Electronics machine learning authentication natural language understanding big data chatbots |
| title | Smart Chatbot for User Authentication |
| title_full | Smart Chatbot for User Authentication |
| title_fullStr | Smart Chatbot for User Authentication |
| title_full_unstemmed | Smart Chatbot for User Authentication |
| title_short | Smart Chatbot for User Authentication |
| title_sort | smart chatbot for user authentication |
| topic | machine learning authentication natural language understanding big data chatbots |
| url | https://www.mdpi.com/2079-9292/11/23/4016 |
| work_keys_str_mv | AT petervoege smartchatbotforuserauthentication AT imanimabusulayman smartchatbotforuserauthentication AT abdelkaderouda smartchatbotforuserauthentication |