Details
Original language | English |
---|---|
Title of host publication | IJCCI 2018 |
Subtitle of host publication | Proceedings of the 10th International Joint Conference on Computational Intelligence |
Editors | Juan Julian Merelo, Christophe Sabourin, Kurosh Madani, Kevin Warwick, Alejandro Linares Barranco |
Pages | 5-14 |
Number of pages | 10 |
ISBN (electronic) | 9789897583278 |
Publication status | Published - 2018 |
Event | 10th International Joint Conference on Computational Intelligence, IJCCI 2018 - Seville, Spain Duration: 18 Sept 2018 → 20 Sept 2018 Conference number: 10 |
Abstract
Adaptive Psychological Profiling is the process of determining a person’s internal mental state through the analysis of a person’s non-verbal behaviour. Silent Talker is a pioneering psychological profiling system which was developed by experts in behavioural neuroscience and computational intelligence. Designed for use in natural conversation, Silent Talker combines image processing and artificial intelligence to classify multiple visible non-verbal signals of the face during verbal communication to produce an accurate and comprehensive time-based profile of a subject’s psychological state. Silent Talker uses a unique configuration of artificial neural networks, hence, it is difficult to understand how the classification of a person’s behaviour is obtained. New legislation in the form of GDPR, now requires individuals whom are automatically profiled, to have the right to an explanation of how the “machine” reached its decision and receive meaningful information on the logic involved. This is difficult in practice, both from a technical and legal point of view. This paper, uses an application of psychological profiling within a pilot system known as iBorderCtrl, which detects deception through an avatar border guard interview during a travellers pre-registration to demonstrate the challenges faced in trying to obtain explainable decisions from models derived through Computational Intelligence techniques.
Keywords
- Artificial neural networks, Decision trees, GDPR, Psychological profiling
ASJC Scopus subject areas
- Computer Science(all)
- Artificial Intelligence
- Computer Science(all)
- Computational Theory and Mathematics
- Computer Science(all)
- Computer Graphics and Computer-Aided Design
- Computer Science(all)
- Software
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
IJCCI 2018: Proceedings of the 10th International Joint Conference on Computational Intelligence. ed. / Juan Julian Merelo; Christophe Sabourin; Kurosh Madani; Kevin Warwick; Alejandro Linares Barranco. 2018. p. 5-14.
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Adapted Psychological Profiling verses the right to an explainable decision
AU - Crockett, Keeley
AU - Stoklas, Jonathan
AU - O’Shea, James
AU - Krügel, Tina
AU - Khan, Wasiq
N1 - Conference code: 10
PY - 2018
Y1 - 2018
N2 - Adaptive Psychological Profiling is the process of determining a person’s internal mental state through the analysis of a person’s non-verbal behaviour. Silent Talker is a pioneering psychological profiling system which was developed by experts in behavioural neuroscience and computational intelligence. Designed for use in natural conversation, Silent Talker combines image processing and artificial intelligence to classify multiple visible non-verbal signals of the face during verbal communication to produce an accurate and comprehensive time-based profile of a subject’s psychological state. Silent Talker uses a unique configuration of artificial neural networks, hence, it is difficult to understand how the classification of a person’s behaviour is obtained. New legislation in the form of GDPR, now requires individuals whom are automatically profiled, to have the right to an explanation of how the “machine” reached its decision and receive meaningful information on the logic involved. This is difficult in practice, both from a technical and legal point of view. This paper, uses an application of psychological profiling within a pilot system known as iBorderCtrl, which detects deception through an avatar border guard interview during a travellers pre-registration to demonstrate the challenges faced in trying to obtain explainable decisions from models derived through Computational Intelligence techniques.
AB - Adaptive Psychological Profiling is the process of determining a person’s internal mental state through the analysis of a person’s non-verbal behaviour. Silent Talker is a pioneering psychological profiling system which was developed by experts in behavioural neuroscience and computational intelligence. Designed for use in natural conversation, Silent Talker combines image processing and artificial intelligence to classify multiple visible non-verbal signals of the face during verbal communication to produce an accurate and comprehensive time-based profile of a subject’s psychological state. Silent Talker uses a unique configuration of artificial neural networks, hence, it is difficult to understand how the classification of a person’s behaviour is obtained. New legislation in the form of GDPR, now requires individuals whom are automatically profiled, to have the right to an explanation of how the “machine” reached its decision and receive meaningful information on the logic involved. This is difficult in practice, both from a technical and legal point of view. This paper, uses an application of psychological profiling within a pilot system known as iBorderCtrl, which detects deception through an avatar border guard interview during a travellers pre-registration to demonstrate the challenges faced in trying to obtain explainable decisions from models derived through Computational Intelligence techniques.
KW - Artificial neural networks
KW - Decision trees
KW - GDPR
KW - Psychological profiling
UR - http://www.scopus.com/inward/record.url?scp=85059131218&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85059131218
SP - 5
EP - 14
BT - IJCCI 2018
A2 - Merelo, Juan Julian
A2 - Sabourin, Christophe
A2 - Madani, Kurosh
A2 - Warwick, Kevin
A2 - Barranco, Alejandro Linares
T2 - 10th International Joint Conference on Computational Intelligence, IJCCI 2018
Y2 - 18 September 2018 through 20 September 2018
ER -