uu.seUppsala University Publications
Change search
Link to record
Permanent link

Direct link
BETA
Ashcroft, Michael
Publications (9 of 9) Show all publications
Ashcroft, M., Johansson, F., Kaati, L. & Shrestha, A. (2016). Multi-domain alias matching using machine learning. In: Proc. 3rd European Network Intelligence Conference: . Paper presented at ENIC 2016, September 5–7, Wroclaw, Poland (pp. 77-84). IEEE
Open this publication in new window or tab >>Multi-domain alias matching using machine learning
2016 (English)In: Proc. 3rd European Network Intelligence Conference, IEEE, 2016, p. 77-84Conference paper, Published paper (Refereed)
Abstract [en]

We describe a methodology for linking aliases belonging to the same individual based on a user's writing style (stylometric features extracted from the user generated content) and her time patterns (time-based features extracted from the publishing times of the user generated content). While most previous research on social media identity linkage relies on matching usernames, our methodology can also be used for users who actively try to choose dissimilar usernames when creating their aliases. In our experiments on a discussion forum dataset and a Twitter dataset, we evaluate the performance of three different classifiers. We use the best classifier (AdaBoost) to evaluate how well it works on different datasets using different features. Experiments show that combining stylometric and time based features yield good results on our synthetic datasets and a small-scale evaluation on real-world blog data confirm these results, yielding a precision over 95%. The use of emotion-related and Twitter-related features yield no significant impact on the results.

Place, publisher, year, edition, pages
IEEE, 2016
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:uu:diva-306944 (URN)10.1109/ENIC.2016.019 (DOI)000399097600011 ()9781509034550 (ISBN)
Conference
ENIC 2016, September 5–7, Wroclaw, Poland
Available from: 2017-02-02 Created: 2016-11-07 Last updated: 2019-03-22Bibliographically approved
Ashcroft, M., Magnani, M., Vega, D., Montesi, D. & Rossi, L. (2016). Multilayer Analysis of Online Illicit Marketplaces. In: Brynielsson, J Johansson, F (Ed.), 2016 European Intelligence And Security Informatics Conference (EISIC): . Paper presented at Conference on European Intelligence and Security Informatics Conference (EISIC), AUG 17-19, 2016, Uppsala, SWEDEN (pp. 199-199). IEEE
Open this publication in new window or tab >>Multilayer Analysis of Online Illicit Marketplaces
Show others...
2016 (English)In: 2016 European Intelligence And Security Informatics Conference (EISIC) / [ed] Brynielsson, J Johansson, F, IEEE , 2016, p. 199-199Conference paper, Published paper (Refereed)
Place, publisher, year, edition, pages
IEEE, 2016
Series
European Intelligence and Security Informatics Conference, ISSN 2572-3723
National Category
Computer Sciences
Identifiers
urn:nbn:se:uu:diva-346563 (URN)10.1109/EISIC.2016.47 (DOI)000411272300042 ()978-1-5090-2857-3 (ISBN)
Conference
Conference on European Intelligence and Security Informatics Conference (EISIC), AUG 17-19, 2016, Uppsala, SWEDEN
Available from: 2018-03-19 Created: 2018-03-19 Last updated: 2018-03-19Bibliographically approved
Ashcroft, M., Kaati, L. & Meyer, M. (2015). A step towards detecting online grooming: Identifying adults pretending to be children. In: Proc. 5th European Intelligence and Security Informatics Conference: . Paper presented at EISIC 2015, September 7–9, Manchester, UK (pp. 98-104). IEEE Computer Society
Open this publication in new window or tab >>A step towards detecting online grooming: Identifying adults pretending to be children
2015 (English)In: Proc. 5th European Intelligence and Security Informatics Conference, IEEE Computer Society, 2015, p. 98-104Conference paper, Published paper (Refereed)
Abstract [en]

Online grooming is a major problem in todays society where more and more time is spent online. To become friends and establish a relationship with their young victims in online communities, groomers often pretend to be children. In this paper we describe an approach that can be used to detect if an adult is pretending to be a child in a chat room conversation. The approach involves a two step process wherein authors are first classified as being children or adults, and then each child is being examined and false children distinguished from genuine children. Our results show that even if it is hard to separate ordinary adults from children in chat logs it is possible to distinguish real children from adults pretending to be children with a high accuracy. In this paper we will discuss the accuracy of the methods proposed, as well as the features that were important in their success. We believe that this work is an important step towards automated analysis of chat room conversation to detect and possible attempts of grooming. Our approach where we use text analysis to distinguish adults who are pretending to be children from actual children could be used to inform children about the true age of the person that they are communicating. This would be a step towards making the Internet more secure for young children and eliminate grooming.

Place, publisher, year, edition, pages
IEEE Computer Society, 2015
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:uu:diva-272202 (URN)10.1109/EISIC.2015.41 (DOI)000380550100014 ()9781479986514 (ISBN)
External cooperation:
Conference
EISIC 2015, September 7–9, Manchester, UK
Available from: 2015-09-09 Created: 2016-01-12 Last updated: 2018-01-10Bibliographically approved
Ashcroft, M., Kaati, L. & Meyer, M. (2015). Are You Really a Child?: A Machine Learning Approach To Protect Children from Online Grooming. In: Proc. National Symposium on Technology and Methodology for Security and Crisis Management: TAMSEC 2015. Paper presented at TAMSEC 2015, November 24–25, Kista, Sweden.
Open this publication in new window or tab >>Are You Really a Child?: A Machine Learning Approach To Protect Children from Online Grooming
2015 (English)In: Proc. National Symposium on Technology and Methodology for Security and Crisis Management: TAMSEC 2015, 2015Conference paper, Poster (with or without abstract) (Refereed)
Abstract [en]

Online grooming and sexual abuse of children is a major threat towards the security of todays society where more and more time is spent online. To become friends and establish a relationship with their young victims in online communities, groomers often pretend to be children. In this work we describe an approach that can be used to detect if an adult is pretending to be a child in a chat room conversation. Our results show that even if it is hard to separate ordinary adults from children in chat logs it is possible to distinguish real children from adults pretending to be children with a high accuracy.

National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:uu:diva-272244 (URN)
Conference
TAMSEC 2015, November 24–25, Kista, Sweden
Available from: 2016-01-12 Created: 2016-01-12 Last updated: 2018-01-10Bibliographically approved
Ashcroft, M., Fisher, A., Kaati, L., Omer, E. & Prucha, N. (2015). Detecting jihadist messages on twitter. In: Proc. 5th European Intelligence and Security Informatics Conference: . Paper presented at EISIC 2015, September 7–9, Manchester, UK (pp. 161-164). IEEE Computer Society
Open this publication in new window or tab >>Detecting jihadist messages on twitter
Show others...
2015 (English)In: Proc. 5th European Intelligence and Security Informatics Conference, IEEE Computer Society, 2015, p. 161-164Conference paper, Published paper (Refereed)
Abstract [en]

Jihadist groups such as ISIS are spreading online propaganda using various forms of social media such as Twitter and YouTube. One of the most common approaches to stop these groups is to suspend accounts that spread propaganda when they are discovered. This approach requires that human analysts manually read and analyze an enormous amount of information on social media. In this work we make a first attempt to automatically detect messages released by jihadist groups on Twitter. We use a machine learning approach that classifies a tweet as containing material that is supporting jihadists groups or not. Even tough our results are preliminary and more tests needs to be carried out we believe that results indicate that an automated approach to aid analysts in their work with detecting radical content on social media is a promising way forward. It should be noted that an automatic approach to detect radical content should only be used as a support tool for human analysts in their work.

Place, publisher, year, edition, pages
IEEE Computer Society, 2015
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:uu:diva-272203 (URN)10.1109/EISIC.2015.27 (DOI)000380550100027 ()9781479986514 (ISBN)
External cooperation:
Conference
EISIC 2015, September 7–9, Manchester, UK
Available from: 2015-09-09 Created: 2016-01-12 Last updated: 2018-01-10Bibliographically approved
Ashcroft, M. (2013). Performing Decision-Theoretic Inference in Bayesian Network Ensemble Models. In: Jaeger, M; Nielsen, TD; Viappiani, P (Ed.), Twelfth Scandinavian Conference on Artificial Intelligence: . Paper presented at Twelfth Scandinavian Conference on Artificial Intelligence, 20-22 November, 2013, Aalborg, Denmark (pp. 25-34). , 257
Open this publication in new window or tab >>Performing Decision-Theoretic Inference in Bayesian Network Ensemble Models
2013 (English)In: Twelfth Scandinavian Conference on Artificial Intelligence / [ed] Jaeger, M; Nielsen, TD; Viappiani, P, 2013, Vol. 257, p. 25-34Conference paper, Published paper (Refereed)
Abstract [en]

The purpose of this paper is to present a simple extension to an existing inference algorithm on influence diagrams (i.e. decision theoretic extensions to Bayesian networks) that permits these algorithms to be applied to ensemble models. The extension, though simple, is important because of the power and robustness that such ensemble models provide [1]. This paper is intended principally as a 'recipe' that can be used even by those unfamiliar with the algorithms extended. Accordingly, I present the algorithms that the original contribution builds upon in full, though references are given to less concise renditions. Those familiar with these algorithms are invited to skip the elucidation. The consequence is a useful paper with more background and less original input than usual.

Series
Frontiers in Artificial Intelligence and Applications, ISSN 0922-6389 ; 257
Keywords
bayesian model averaging, influence diagrams, probabilistic models, graphical models, artificial intelligence, machine learning
National Category
Computer Sciences
Identifiers
urn:nbn:se:uu:diva-211926 (URN)10.3233/978-1-61499-330-8-25 (DOI)000343477100004 ()978-1-61499-329-2 (ISBN)
Conference
Twelfth Scandinavian Conference on Artificial Intelligence, 20-22 November, 2013, Aalborg, Denmark
Available from: 2013-12-03 Created: 2013-12-03 Last updated: 2018-01-11Bibliographically approved
Ashcroft, M. (2012). An Introduction To Bayesian Networks in Systems and Control. Paper presented at 2012 18th International Conference on Automation and Computing (ICAC).
Open this publication in new window or tab >>An Introduction To Bayesian Networks in Systems and Control
2012 (English)Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:uu:diva-185566 (URN)
Conference
2012 18th International Conference on Automation and Computing (ICAC)
Available from: 2012-11-26 Created: 2012-11-26 Last updated: 2018-01-12
Ashcroft, M. (2012). Bayesian Networks in Business Analytics. In: 2012 FEDERATED CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS (FEDCSIS). Paper presented at Federated Conference On Computer Science And Information Systems; 9 - 12 September, 2012; Wrocław, Poland (pp. 955-961).
Open this publication in new window or tab >>Bayesian Networks in Business Analytics
2012 (English)In: 2012 FEDERATED CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS (FEDCSIS), 2012, p. 955-961Conference paper, Published paper (Refereed)
National Category
Computer Sciences
Identifiers
urn:nbn:se:uu:diva-185567 (URN)000312714400134 ()978-83-60810-48-4 (ISBN)
Conference
Federated Conference On Computer Science And Information Systems; 9 - 12 September, 2012; Wrocław, Poland
Available from: 2012-11-26 Created: 2012-11-26 Last updated: 2018-01-12Bibliographically approved
Ashcroft, M. (2012). Using Bayesian networks in business analytics: Overview and short case study. Business Informatics, 3(25)
Open this publication in new window or tab >>Using Bayesian networks in business analytics: Overview and short case study
2012 (English)In: Business Informatics, ISSN 1507-3858, Vol. 3, no 25Article in journal (Refereed) Published
National Category
Computer Sciences
Identifiers
urn:nbn:se:uu:diva-188761 (URN)
Available from: 2012-12-19 Created: 2012-12-19 Last updated: 2018-01-11Bibliographically approved
Organisations

Search in DiVA

Show all publications