Software Engineering Laboratory

Minetada Osano Professor	Vitaly Klyuev Associate Professor

Information Retrieval Direction Cross-lingual information retrieval and text mining are key topics in research conducted by the Software Engineering Lab members. This year, the focus was on techniques related to Web search technologies applied to Japanese and Chinese. Results of our investigations and tests were presented at the NTCIR 6 Workshop. The aim of our work in the area of text mining was to improve Web search service. The retrieval efficiency of the presently used search tools cannot be made better: A \bag of words" interpretation causes loosing semantics of texts. The functional approach to present English texts in the memory of computers makes it possible to keep semantic relations between words and use ordinary English sentences as queries. The prototype of the system utilizing this approach was proposed. Our study was supported by a grant from the Japan Science and Technology Agency, and a grant from the University of Aizu Competitive Research Funding. Students of our lab were involved in this research. Results of our study were presented at the 7th IEEE Conference on Computer and Information Technology (CIT 2007), the 37th Annual IEEE Frontiers in Education Conference (FIE 2007), and at The 10th International Conference on Humans and Computers (HC 2007).

Program for Leading Edge IT Specialists Our lab joined to the Program for Leading Edge IT Specialists organized by our university and offered students a project on Summarizing Tools for General Purpose Search Engines. A team-based way of education and research is a new style of teaching for us. We are working now with the rst international group of students consisting of Japanese and Russian students.

International Relations International Project on Text Mining This year, we worked in tight cooperation with our Russian partners (Faculty of Applied Mathematics and Control Processes Saint Petersburg State University) on the international project on Text Mining. Results of our joint research were presented at the 37th Annual IEEE Frontiers in Education Conference (FIE 2007). Exchange of undergraduate students Our undergraduate student Mr. Nobuoka visited Saint Petersburg State University, Russia in April 2007 and presented his paper at the XXXVIII Conference on Control Processes and Stability. Russian undergraduate student Mr. Lipenkov attended the HC 2007 Conference in December 2007. This exchange of students was done in accordance with our agreement with Saint Petersburg State University.

Refereed Journal Papers

[vkluev-01:2007]	M. Mozgovoy, V. Tuzov, and V. Klyuev. A Fast Semantic-Powered Plagiarism Detection system. The Journal of Three Dimentional Images, 21(2):105 - 110, 2007.
	Plagiarism detection systems are known for years in the university community. However, most of the existing detectors for the natural language texts use rather simple comparison methods that make the instances of plagiarism easy to hide. The software, designed for plagiarism detection in computer programs, utilizes far more advanced techniques. We propose a method, which adds functionalities similar to tokenization and tree matching, to the natural language texts-oriented detectors. This method requires noticeable work to be applied in practice, but also makes use of the existing software for parsing and word sense disambiguation.
[vkluev-02:2007]	V. Klyuev, T.Tsushimoto, and G. Nikishkov. Using a Web-Based System to support Teaching Processes. International Journal of Information and Communication Technology Education, 4(1):72 - 85, 2008.
	A platform-independent Java Web application named TSI (Teacher-Student Interaction) that supports communication between an instructor, teaching assistants, and students in a traditional on-campus course is presented in this article. Using the TSI, the instructor and teaching assistants can handle most of the routine work: upload student personal information, send students personal e-mails, etc. The system can easily be installed and administered individually by an instructor inexperienced in computers. It is as simple as a pen for students. Students can check their personal data (scores and comments), download educational materials, etc. As part of the TSI, a VBA application is used to analyze the course log les. This tool is helpful in understanding individual and group studentsbehaviors. The TSI was successfully tested during four years at the University of Aizu (Japan) in an environment where English is one of the working languages and both students and professors are non-native speakers of English.

Refereed Proceeding Papers

[vkluev-03:2007]	T. Saito and V. Klyuev. Comparison of the Index Size of Google, Yahoo!, Ask, and MSN. In Proceedings of the Tenth International conference on Humans and Computers, pages 196 - 199. the University of Aizu, the University of Aizu, Dec. 2007.
	We applied the technique proposed by M. Chency and M. Perry to evaluate the index size of Google, Yahoo!, MSN ans Ask.
[vkluev-04:2007]	K. Sato and V. Klyuev. Finding Advertising Keywords on Japanese Web Pages. In Proceedings of the Tenth International conference on Humans and Computers, pages 187 - 189. the University of Aizu, the University of Aizu, Dec. 2007.
	We investigated several aspects for appropriate keyword extraction from Japanese Web pages.
[vkluev-05:2007]	E. Nagaoka and V. Klyuev. An On-line Directory for Students. In Proceedings of the Tenth International conference on Humans and Computers, pages 190 - 191. the University of Aizu, the University of Aizu, Dec. 2007.
	We propose several improvements for the recenntly introduced tools such as Google Notebook by Google and My Stuff by Ask3D.
[vkluev-06:2007]	V. Klyuev and V. Oleshchuk. Semantic Retrieval of Text Documents. In 7th IEEE International Conference on Computer and Information Technology (CIT 2007), pages 189 - 193. IEEE, IEEE Computer Society, Oct. 2007.
	Nowadays, the Internet is the major source of information for millions of people. There are many search tools available on the net but nding appropriate text information is still difficult. The retrieval effciency of the presently used systems cannot be significantly improved:�@�B ag of wordsinterpretation causes loosing semantics of texts. We applied the functional approach to present English text documents in the memory of computers. It allowed to keep semantic relations between words when indexing documents and use ordinary English sentences as queries to submit to a search engine. The proposed retrieval algorithm returns highly relevant documents. The presented example illustrates the advantage of the discussed approach compared to the traditional key word search.
[vkluev-07:2007]	M. Mozgovoy, S. Karakovsky, and V. Klyuev. Fast and Reliable Plagiarism Detection System. In 37th Annual IEEE Frontiers in Education Con- ference. IEEE, IEEE Computer Society, Oct. 2007.
	Plagiarism and similarity detection software is well-known in universities for years. Despite the variety of methods and approaches used in plagiarism detection, the typical trade-off between the speed and the reliability of the algorithm still remains. We introduce a new two-step approach to plagiarism detection that combines high algorithmic performance and the quality of pairwise file comparison. Our system uses fast detection method to select suspicious files only, and then invokes precise (and slower) algorithms to get reliable results. We show that the proposed method does not noticeably reduce the quality of the pairwise comparison mechanism while providing better speed characteristics.
[vkluev-08:2007]	V. Klyuev. OASIS at NTCIR-6: On-line Query Translation for Chinese- Japanese Cross-Lingual Information Retrieval. In NTCIR Workshop 6 Meet- ing on Evaluation of Information Access Technologies: Information Retrieval, Question Answering and Cross-Lingual Information Access, pages 85 - 91, Tokyo, May 2007. National Institute of Informatics, National Institute of Informatics.
	The aim of this study is to investigate the effciency of on-line translation systems in managing the polysemy problem for the query terms, when systems automatically translate them. This paper reports results of Chinese Japanese CLIR experiments using on-line query translation techniques. Approaches to use English as a pilot language and several on-line translation systems are introduced. They were tested using NTCIR-3, 4, 5, and 6 collections. They can be helpful under certain circumstances.
[vkluev-09:2007]	M. Sampei and V. Klyuev. Finding Stop Words for Japanese. In Proceedings of the Tenth International conference on Humans and Computers, pages 192 - 195. the University of Aizu, the University of Aizu, Dec. 2007.
	In this study, we applied a statistical approach to find stop word for Japanese

Unrefereed Papers

[vkluev-10:2007]

V. Klyuev. Using English for Queries: An Approach to Implementing an Intelligent Web Search. In IPSJ SIG-DD/FI, pages 47-51, Tokyo, Mar. 2008. IPSJ, IPSJ.

The retrieval efficiency of the presently used search tools cannot be signicantly improved: A 'bag of words' interpretation causes loosing semantics of texts. The functional approach to present English texts in the memory of computers makes it possible to keep semantic relations between words and use ordinary English sentences as queries. The prototype of the system utilizing this approach is presented.

Grants

[vkluev-11:2007]	V. Klyuev. A grant from the Japan Science and Technology Agency, 2007.
[vkluev-12:2007]	V. Klyuev. a grant from the University of Aizu Competitive Research Funding, 2007.

Academic Activities

[vkluev-13:2007]

V. Klyuev, Apr. 2007. Member, IEEE, ACM

Ph.D, Master and Graduation Theses

[vkluev-14:2007]	Emu Nagaoka. Graduation Thesis: Comparing Online Directories for Students, University of Aizu, 2008.
	Thesis Advisor: Klyuev, V
[vkluev-15:2007]	Keiuske Sato. Graduation Thesis: Finding Advertising Keywords on Japanese Web Pages, University of Aizu, 2008.
	Thesis Advisor: Klyuev, V
[vkluev-16:2007]	Masashiro Sampei. Graduation Thesis: Finding Stop Words for Japanese, University of Aizu, 2008.
	Thesis Advisor: Klyuev, V
[vkluev-17:2007]	Tatsuya Saito. Graduation Thesis: Comparison of the Index Size of Google, Yahoo!, Ask, and MSN, University of Aizu, 2008.
	Thesis Advisor: Klyuev, V

2009 THE UNIVERSITY OF AIZU ALL RIGHTS RESERVED