Evaluation of authorship attribution software on a Chat bot corpus

Ali, Nawaf; Hindi, Musa; Yampolskiy, Roman V.

doi:10.1109/icat.2011.6102123

Cited by 18 publications

(14 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In such scenarios, it becomes important to be able to detect changes in the agent's reward function caused by forced or self-administered wireheading. Behavioural profiling of artificially intelligent agents may present a potential solution to wireheading detection (Ali, Hindi, & Yampolskiy, 2011;, 16-20 March 2008.…”

Section: Discussionmentioning

confidence: 99%

Utility function security in artificially intelligent agents

Yampolskiy

2014

Journal of Experimental & Theoretical Artificial Intelligen

Self Cite

View full text Add to dashboard Cite

The notion of 'wireheading', or direct reward centre stimulation of the brain, is a wellknown concept in neuroscience. In this paper, we examine the corresponding issue of reward (utility) function integrity in artificially intelligent machines. We survey the relevant literature and propose a number of potential solutions to ensure the integrity of our artificial assistants. Overall, we conclude that wireheading in rational selfimproving optimisers above a certain capacity remains an unsolved problem despite opinion of many that such machines will choose not to wirehead. A relevant issue of literalness in goal setting also remains largely unsolved and we suggest that the development of a non-ambiguous knowledge transfer language might be a step in the right direction.

show abstract

Section: Discussionmentioning

confidence: 99%

Utility function security in artificially intelligent agents

Yampolskiy

2014

Journal of Experimental & Theoretical Artificial Intelligen

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, Sidorov et al (2014) conducted an authorship attribution study using a set of three authors of 39 literary texts. In another study, Ali et al (2011) examined methods for identification of chat bots using a set of 11 candidate authors. Even though the authorial pool or the corpus size is considered small, the feature space may be measured in thousands, because feature extraction techniques are only limited by computational capacity and the researcher's approach.…”

Section: Methodsmentioning

confidence: 99%

Using Learning Analytics for Preserving Academic Integrity

Amigud¹,

Arnedo-Moreno²,

Daradoumis³

et al. 2017

IRRODL

View full text Add to dashboard Cite

This paper presents the results of integrating learning analytics into the assessment process to enhance academic integrity in the e-learning environment. The goal of this research is to evaluate the computational-based approach to academic integrity. The machine-learning based framework learns students' patterns of language use from data, providing an accessible and non-invasive validation of student identities and student-produced content. To assess the performance of the proposed approach, we conducted a series of experiments using written assignments of graduate students. The proposed method yielded a mean accuracy of 93%, exceeding the baseline of human performance that yielded a mean accuracy rate of 12%. The results suggest a promising potential for developing automated tools that promote accountability and simplify the provision of academic integrity in the e-learning environment.

show abstract

“…Author identification is a form of biometric recognition [23]. There are four primary components to any biometric system: a sensor module, a feature extraction module, a matching module, and a database module [15].…”

mentioning

confidence: 99%

Genetic Heuristic Development: Feature selection for author identification

Adams

Williams

Carter

et al. 2013

2013 IEEE Symposium on Computational Intelligence in Biometrics and Identity Management (CIBIM)

View full text Add to dashboard Cite

Author identification is the process of recognizing an author based on a sample of text. Feature selection is the process of selecting the most salient features required for recognition. In many cases, this results in an increase in recognition accuracy. In this paper, we apply Genetic and Evolutionary Feature Selection with Machine Learning (GEFeS ML ) to author identification. We then introduce Genetic Heuristic Development (GHD), a process to improve the matching process. GHD uses subsets of features found by GEFeS ML to create a high performing heuristic for feature selection.This technique successfully increases recognition accuracy while significantly reducing the number of features required for recognition.

show abstract

Evaluation of authorship attribution software on a Chat bot corpus

Cited by 18 publications

References 6 publications

Utility function security in artificially intelligent agents

Utility function security in artificially intelligent agents

Using Learning Analytics for Preserving Academic Integrity

Genetic Heuristic Development: Feature selection for author identification

Contact Info

Product

Resources

About