7th International Conference on Natural Language Processing. Published by Springer.

Eriks Sneiders. Automated Email Answering by Text Pattern Matching.
Abstract. Answering email by standard answers is a common practice at contact centers. Our research assists this process by creating reply messages that contain one or several standard answers. Our standard answers are linked to representative text patterns that match incoming messages. The system works in three languages. The performance was evaluated on two email sets; the main advantage of our email answering technique is good correctness of the delivered replies.

Hercules Dalianis, Magnus Rosell, Eriks Sneiders. Clustering E-Mails for the Swedish Social Insurance Agency? What Part of the E-Mail Thread Gives the Best Quality?
Abstract. We need to analyze a large amount of emails sent to a customer service of a governmental organization based in Sweden from citizens. To carry out this analysis we need to cluster a large number of emails with the aim of automatic email answering. The research problem is whether we should use the whole email including the thread or just the original query for the clustering. We have in this paper described the experiments to find out this. Our results shows that only the query and the answering part should be used, but not necessarily the whole email thread. Our normalized mutual information (NMI) for the query and answering part is 0.12 higher than for only the query.

7th International Conference on Natural Language Processing. Published by Springer.

Eriks Sneiders. Automated Email Answering by Text Pattern Matching.
Abstract. Answering email by standard answers is a common practice at contact centers. Our research assists this process by creating reply messages that contain one or several standard answers. Our standard answers are linked to representative text patterns that match incoming messages. The system works in three languages. The performance was evaluated on two email sets; the main advantage of our email answering technique is good correctness of the delivered replies.

Hercules Dalianis, Magnus Rosell, Eriks Sneiders. Clustering E-Mails for the Swedish Social Insurance Agency? What Part of the E-Mail Thread Gives the Best Quality?
Abstract. We need to analyze a large amount of emails sent to a customer service of a governmental organization based in Sweden from citizens. To carry out this analysis we need to cluster a large number of emails with the aim of automatic email answering. The research problem is whether we should use the whole email including the thread or just the original query for the clustering. We have in this paper described the experiments to find out this. Our results shows that only the query and the answering part should be used, but not necessarily the whole email thread. Our normalized mutual information (NMI) for the query and answering part is 0.12 higher than for only the query.

October 27, 2009 DSV submitted a project proposal to Information and Communication Technologies Calls: FP7-ICT-2009-5 . Eriks Sneiders was the main author and the coordinator. Participants:
– Fraunhofer-Institut für Software- und Systemtechnik (Germany)
– Stockholms universitet (Sweden)
– Health on the Net (Switzerland)
– Centre National de la Recherche Scientifique (France)
– Sabiedr?ba ar ierobežotu atbild?bu Tilde (Latvia)
– Uniwersytet Ekonomiczny w Poznaniu (Poland)
– T-Systems (Germany)
– BARMER Ersatzkasse (Germany)

Abstract
The Internet is known to be an outstanding tool for acquiring information and services. However, there exist three gaps that hamper the process of content acquisition performed by users. The first gap is the accessibility of buried content connected with the length of the path users have to follow to acquire the content. The second gap is the people’s inability to use existing tools to access content caused by cognitive discrepancy between developed tools and users’ perception and preferences. And finally, information overload caused by large amount of information, less time to process information, and complexity and ambiguity of information.

The goal of the AQUA project is to reduce all three gaps. The project will deliver technologies that will provide the shortest possible path between users information needs and the content (by providing high quality answers to questions and advancing cross-lingual question answering research area), take into account cognitive characteristics of users (by developing a user-centred knowledge representation and using it within the information delivery process) and will shield a user from information overload (by developing user-centred and context-based information supply mechanisms).

AQUA will use medical information as a test bed as this is a domain where: people need exact and reliable information; there are a huge number of adequate and inadequate information sources on the Internet providing lots of data, and finally, where faulty or incomplete information may have serious consequences for the individual. The content for the AQUA system will be inter alia HONcode members (6 800 health sites in 72 countries) approved for medical accuracy by Health on the Net (HON) as well as medical insurance data.

October 27, 2009 DSV submitted a project proposal to Information and Communication Technologies Calls: FP7-ICT-2009-5 . Eriks Sneiders was the main author and the coordinator. Participants:
– Fraunhofer-Institut für Software- und Systemtechnik (Germany)
– Stockholms universitet (Sweden)
– Health on the Net (Switzerland)
– Centre National de la Recherche Scientifique (France)
– Sabiedr?ba ar ierobežotu atbild?bu Tilde (Latvia)
– Uniwersytet Ekonomiczny w Poznaniu (Poland)
– T-Systems (Germany)
– BARMER Ersatzkasse (Germany)

Abstract
The Internet is known to be an outstanding tool for acquiring information and services. However, there exist three gaps that hamper the process of content acquisition performed by users. The first gap is the accessibility of buried content connected with the length of the path users have to follow to acquire the content. The second gap is the people’s inability to use existing tools to access content caused by cognitive discrepancy between developed tools and users’ perception and preferences. And finally, information overload caused by large amount of information, less time to process information, and complexity and ambiguity of information.

The goal of the AQUA project is to reduce all three gaps. The project will deliver technologies that will provide the shortest possible path between users information needs and the content (by providing high quality answers to questions and advancing cross-lingual question answering research area), take into account cognitive characteristics of users (by developing a user-centred knowledge representation and using it within the information delivery process) and will shield a user from information overload (by developing user-centred and context-based information supply mechanisms).

AQUA will use medical information as a test bed as this is a domain where: people need exact and reliable information; there are a huge number of adequate and inadequate information sources on the Internet providing lots of data, and finally, where faulty or incomplete information may have serious consequences for the individual. The content for the AQUA system will be inter alia HONcode members (6 800 health sites in 72 countries) approved for medical accuracy by Health on the Net (HON) as well as medical insurance data.

Eriks Sneiders held a lunch seminar about his research in automated question and e-mail answering. The research has lead to real life systems, such as a QA engine for Trygg-Hansa and e-mail answering engine for Gjensidige Försäkring in Sweden. The systems demonstrate rather high performance figures, thanks to granularity of the knowledge representation.

Eriks Sneiders held a lunch seminar about his research in automated question and e-mail answering. The research has lead to real life systems, such as a QA engine for Trygg-Hansa and e-mail answering engine for Gjensidige Försäkring in Sweden. The systems demonstrate rather high performance figures, thanks to granularity of the knowledge representation.

It does not happen often that a master thesis becomes a conference paper, but it does happen. Marjan Khatir, an EMIS student, has got her paper “How to Find Exculpatory and Inculpatory Evidence Using a Circular Digital Forensics Process Model” accepted at the 4th International Conference on Global E-Security. Congratulations, Marjan!

Abstract
With raising the number of cyber crimes, the need of having a proper digital forensic process also increases. Although digital forensics is practiced in recent years, there is still a big gap between previously suggested digital forensics processes and what is really needed to be done in real cases. Some problems with current processes are lack of flexible transition between phases, not having a clear method or a complete scenario for addressing reliable evidence, and not paying enough attention to management aspects and team roles. This paper provides a process model by paying special attention to the team roles and management aspects as well as both exculpatory and inculpatory evidence.