![](bg6.png)
representation of the similarity relations between the
articles. Advantages of this hierarchy include
logarithmical complexity, metadata which are
generated using content of the articles and
incremental approach. This is useful if we need real-
time calculation and the metadata provided by the
authors of the news are not sufficient. On the other
hand, a disadvantage is that the tree structure could
not provide relations which are not transitive (i.e.
text similarity of news).
We use properties of the hierarchical
representation in our method. The results thus meet
the requirements of the recommender system.
Hierarchical clustering has low, logarithmical
complexity of storing and retrieving articles. The
hierarchy enables us to discover interests for every
moment using the history of reading.
Our main contribution is utilization of
hierarchical structure, which incrementally generates
metadata about articles. Meta-documents which are
created this way have inheritance relations. These
relations represent similarity between real articles.
The advantages of our recommender systems are
linked to this representation. We are able to discover
user’s interests in real-time, even if we use vast
information space to recommend news.
We focused in our work on real-time content-
based recommending. Our future work includes
considering the context of the user’s interests. We
plan to improve our recommender to consider the
actual interests of a user. We have a presumption
that interests change in time, with location, mood or
emotions. Since we are able to recommend news in
real-time, this is mainly a matter of recognizing the
behavioural patterns and contexts.
ACKNOWLEDGEMENTS
This work was supported by the Scientific Grant
Agency of SR, grants No. VG1/0508/09 and
VG1/0675/11, and it is a partial result of the
Research & Development Operational Program for
the project Support of Center of Excellence for
Smart Technologies, Systems and Services II, ITMS
25240120029, co-funded by ERDF.
REFERENCES
Ahn, J., Brusilovsky, P., Grady, J., He, D., and Syn, S. Y.
2007. Open user profiles for adaptive news systems:
help or harm?. In Proc. of the 16th int. Conf.on World
Wide Web. WWW '07. ACM, New York, NY, 11-20.
Adomavicius, G. and Tuzhilin, A. 2005. Toward the Next
Generation of Recommender Systems. IEEE Trans. on
Knowl. and Data Eng. 17, 6, 734-749.
Barla, M. et al., 2010. News recommendation. In Proc. of
the 9th Znalosti, Jindrichuv Hradec., 171-174.
Billsus, D., Pazzani, M. 2000. User Modeling for Adaptive
News Access. User Modeling and User-Adapted
Interaction, vol. 10, nos. 2-3, (Feb. 2000),147-180.
Bollen, D., Knijnenburg, B. P., & Graus, M. 2010.
Understanding Choice Overload in Recommender
Systems Categories and Subject Descriptors. In Proc.
of 4th ACM Conf. on Recommender Systems.
Barcelona, Spain, 63-70.
Burke, R. 2002. Hybrid Recommender Systems: Survey
and Experiments. User Modeling and User-Adapted
Interaction 12, 4 (Nov. 2002), 331-370.
Carvalho, C., Jorge, A. M., and Soares, C. 2006.
Personalization of E-newsletters Based on Web Log
Analysis and Clustering. In Proc. of the
IEEE/WIC/ACM Int. Conf. on Web intelligence. IEEE
Computer Society, WDC, 724-727.
Gabrilovich, E., Markovitch, S. 2007. Computing
semantic relatedness using Wikipedia-based explicit
semantic analysis. In Proc. of the 20th int. Joint Conf.
on Artificial Intelligence, Hyderabad., India, 1606-
1611.
Ge, M., Delgado-battenfeld, C. 2010. Beyond Accuracy:
Evaluating Recommender Systems by Coverage and
Serendipity. In Proc. of 4th ACM Conf. on
Recommender. Systems. Barcelona, Spain, 257-260.
Jancsary, J., Neubarth, F., Trost, H. 2010. Towards
Context-Aware Personalization and a Broad
Perspective on the Semantics of News Articles. In
Proc. of 4th ACM Conf. on Recommender Systems,
Barcelona, Spain, 289-292.
Kroha, P.,Baeza-Yates, R., 2005. News classification
based on term frequency. In Proc. of the 16th Conf. on
Database and Expert Sys. Apps, 428–432.
Kompan, M., Bieliková, M., 2010. Content-Based News
Recommendation. In Proc. of the 11th Conf. EC-WEB.
Springer-Verlag, Bilbao, Spain, 61-72.
Mooney, R. J. and Roy, L. 2000. Content-based book
recommending using learning for text categorization.
In Proc. of the 5th Conf. on Digital Libraries, TX,
USA, 195-204.
Sahoo, N., Callan, J., Krishnan, R., Duncan, G., Padman,
R. 2006. Incremental hierarchical clustering of text
documents. In Proc. of the 15th ACM int. Conf. on
Information and knowledge management, NY, USA,
357-366.
Su, X. and Khoshgoftaar, T. M. 2009. A survey of
collaborative filtering techniques. Adv. in AI, 36-55.
Suchal, J., Navrat, P. 2010. Full text search engine as
scalable k-nearest neighbor recommendation system.
In Proc. of the AI in Theory and Practice 2010. WCC.
IFIP AICT 331, Springer, Boston, 165-173.
Tintarev, N., Masthoff, J. 2006. Similarity for news
recommender systems. In Proc. of the AH’06
Workshop on Recommender Systems and Intelligent
User Interfaces, 1-8.
NEWS RECOMMENDING BASED ON TEXT SIMILARITY AND USER BEHAVIOUR
307