Wilhelmiina Hämäläinen
Ruuska, S., Hämäläinen, W., Kajava, S., Mughal, M., Matilainen, P., Mononen, J.: Evaluation of the confusion matrix method in the validation of an automated system for measuring feeding behaviour of cattle. Behavioural Processes 148: 56-62, 2018. publisher's link
Hämäläinen, W. and Webb, G.I.: Specious rules: an efficient and effective unifying method for removing misleading and uninformative patterns in association rule mining. Proceedings of SIAM International Conference on Data Mining, pp. 309-317, SIAM 2017. paper
Hämäläinen, W., Ruuska, S., Kokkonen, T., Orkola, S., Mononen, J.: Measuring behaviour accurately with instantaneous sampling: A new tool for selecting appropriate sampling intervals. Applied Animal Behaviour Science 180:166-73, 2016. publisher's link
Hämäläinen, W.: New upper bounds for tight and fast approximation of Fisher's exact test in dependency rule mining. Computational Statistics & Data Analysis 93:469-482, 2016. publisher's link
Hämäläinen, W. and Webb, G.I.: Statistically sound pattern discovery (Tutorial summary paper). Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'14). Page 1976, ACM 2014.
Ruuska, S., Hämäläinen, W., Sairanen, A., Juutinen, E., Tuomisto, L., Järvinen, M. and Mononen, J.: Can stealing cows distort the results of feeding trials? An experiment for quantification and prevention of stealing feed by dairy cows from roughage intake control feeders. Applied Animal Behaviour Science 159:1-8, 2014. publisher's link
Hämäläinen, W., Petitjean, F. and Webb, G.I. (editors): Proceedings of the 1st ECML/PKDD 2014 workshop on Statistically Sound Data Mining. JMLR Workshop and Conference Proceedings series, volume 47, JMLR, 2015. proceedings
S. Ruuska, W. Hämäläinen, T. Kokkonen, S. Orkola and J. Mononen: Reliability of instantaneous sampling in measuring feeding behaviour of dairy cows. Proceedings of the 25th Nordic Symposium of the International Society for Applied Ethology, p. 35, 2014.
Hämäläinen, W., V. Kumpulainen, M. Mozgovoy: Evaluation of clustering methods for adaptive learning systems. In U. Käse and D. Koc (editors): Artificial Intelligence Applications in Distance Education. Pages 237-260. IGI Global, 2014. own version (notice: different page numbers and layout than in the book)
Hämäläinen, W.: Kingfisher: an efficient algorithm for searching for both positive and negative dependency rules with statistical significance measures. Knowledge and Information Systems: An International Journal (KAIS) 32(2):383-414, 2012. Pre-publication version Springer link bibtex
Hämäläinen, W.: Thorough analysis of log data with dependency rules: Practical solutions and theoretical challenges. Proceedings of Data Mining Workshops (ICDMW), 2012 IEEE 12th International Conference on. Pages 579 - 586. IEEE, 2012. prepublication version
Hämäläinen, W.: Efficient search methods for statistical dependency rules. Fundamenta Informaticae (Special issue on Statistical and Relational Learning in Bioinformatics) 113(2):117-150, 2011. Pre-publication version bibtex
W. Hämäläinen, M. Järvinen, P. Martiskainen, J. Mononen: Jerk-based Feature Extraction for Robust Activity Recognition from Acceleration Data. Proceedings of the 11th International Conference on Intelligent Systems Design and Applications (ISDA 2011). Pages 831-836. IEEE Computer Society 2011. Pre-publication version bibtex
Hämäläinen, W.: Efficient discovery of the top-K optimal dependency rules with Fisher's exact test of significance. Proceedings of the 10th IEEE International Conference on Data Mining (ICDM 2010), pp. 196-205. IEEE Computer Society 2010. Pre-publication version IEEE link slides (contains a step-by-step simulation of the algorithm!) bibtex
Hämäläinen, W.: StatApriori: an efficient algorithm for searching statistically significant association rules. Knowledge and Information Systems: an International Journal (KAIS) 23(3): 373-399, 2010. source code bibtex Pre-publication version Springer link
Hämäläinen, W. and Vinni, M.: Classifiers for educational data mining. In Romero, C., et al., editors, Handbook on Educational Data Mining. Pages 57--74. CRC Press. Chapman & Hall/CRC Data Mining and Knowledge Discovery Series, 2010. own version (different page numbers and layout than in the book)
W. Hämäläinen, P. Martiskainen, M. Järvinen, J.-P. Skön, J. Tiirikainen, M. Kolehmainen and J. Mononen: Computational challenges in deriving dairy cows' action patterns from accelerometer data. Proceedings of the 22nd Nordic Symposium of the International Society for Applied Ethology, p. 18, 2010. presentation.
Hämäläinen, W.: Lift-based search for significant dependencies in dense data sets. Proceedings of the Workshop on Statistical and Relational Learning in Bioinformatics (StReBio'09), in the 15th ACM SIGKDD conference on Knowledge Discovery and Data Mining, pp. 12-16, ACM, 2009. bibtex slides
Hämäläinen, W. and Nykänen, M.: Efficient discovery of statistically significant association rules. Proceedings of the 8th IEEE International Conference on Data Mining (ICDM 2008), pp. 203-212. IEEE Computer Society 2008. bibtex slides
Hämäläinen, W. and Vinni, M.: Comparison of machine learning methods for intelligent tutoring systems. Proceedings of the 8th international conference in intelligent tutoring systems (ITS'06), LNCS 4053. Pages 525-534. Springer, 2006. bibtex
Hämäläinen, W., Laine, T.H. and Sutinen, E.: Data mining in personalizing distance education courses. In C. Romero and S. Ventura, editors, Data Mining in E-learning. Pages 157-171. WitPress, Southampton, 2005. bibtex
Hämäläinen, W.: General paradigms for implementing adaptive learning systems. Proceedings of the IADIS multi conference on computer science and information systems (MCCSIS 2005). Pages 476-483. 2005. bibtex
Hämäläinen, W., Poroshin, V. and Toivonen, H.: Mining relaxed graph properties in Internet. Proceedings of IADIS WWW/Internet 2004. Pages 152-159. bibtex
Hämäläinen, W., Suhonen, J., Sutinen, E. and Toivonen, H. Data Mining in Personalizing Distance Education Courses 21st ICDE World Conference on Open Learning and Distance Education, Hong Kong, February 18-21, 2004. Note: the paper was accepted and presented but not published in the proceedings due to a technical error. bibtex
Hämäläinen, W.: Problem-based learning of theoretical computer science. Proceedings of the 34th Annual Conference on Frontiers in Education (FIE'2004). Pages S1H/1-S1H/6 Vol. 3. IEEE, 2004.
Hämäläinen, W.: Statistical analysis of problem-based learning in theory of computation. Kolin Kolistelut - Koli Calling 2004, Proceedings of the Fourth Finnish/Baltic Sea Conference on Computer Science Education. pp. 101-106.
Hämäläinen, W. Problem-based learning of theoretical computer science. Kolin Kolistelut - Koli Calling 2003, Proceedings of the Third Finnish/Baltic Sea Conference on Computer Science Education. pp. 50-58.
Hämäläinen, W. Teaching Computer Science by Playing. Kolin Kolistelut - Koli Calling 2002, Proceedings of the Second Finnish/Baltic Sea Conference on Computer Science Education. Pages 43-49.
Hämäläinen, W., Myller, N., Lopez-Cuadrado, J. and Pitkänen, S. A Self-Motivating Adaptive Learning System. Kolin Kolistelut - Koli Calling 2002, Proceedings of the Second Finnish/Baltic Sea Conference on Computer Science Education. Pages 22-27.
Hämäläinen, W., Webb, G.: A Tutorial on Statistically Sound Pattern Discovery. arXiv:1709.03904 [stat.ME], ArXiv e-prints, September 2017. paper
Hämäläinen, W.: General upper bounds for well-behaving goodness measures on de- pendency rules. arXiv:1405.1339 [cs.DB], ArXiv e-prints, May 2014. paper
Hämäläinen, W.: New tight approximations for Fisher’s exact test. arXiv:1405.1250 [stat.CO], ArXiv e-prints, May 2014. paper
Hämäläinen, W.: Assessing the statistical significance of association rules. arXiv:1405.1360 [cs.DB], ArXiv e-prints, May 2014. paper
Hämäläinen, W.: Efficient search for statistically significant dependency rules in binary data. Department of Computer Science, University of Helsinki, Finland. Series of Publications A, Report A-2010-2. Abstract pdf Corrections and additional notes
Hämäläinen, W.: Descriptive and Predictive Modelling Techniques for Educational Technology. Licentiate thesis. Department of Computer Science, University of Joensuu, Finland. 2006. Abstractpdf bibtex
I have done two interdisiplinary master theses, which deal with both philosophy and computer science - the focus is naturally according to the major degree in question.
Principia conclusionis incertae: Epävarman päättelyn mallintaminen Bayes-verkoilla, Dempster-Shafer -teorialla ja TM-järjestelmillä ("Principia conclusionis incertae: Modelling uncertain reasoning with Bayesian networks, Dempster-Shafer theory and TM-systems"). Master's thesis, Department of Computer Science, University of Helsinki, 2001. brief in English and a related paper
Sielun matka virtuaalitodellisuuteen - virtuaalimaailma mahdollisena maailmana, Popperin maailma 3:na sekä Goodmanin maailmanversiona ("The soul's journey into Virtual Reality - virtual world as posible world, Popper's world 3 and Goodman's world version"). Master's thesis, Department of Systemathic Theology, University of Helsinki, 1998.
Ruuska S, Hämäläinen W, Sairanen A, Juutinen E, Tuomisto L, Järvinen M, and Mononen J. Lypsylehmien rehunvarastamiskäyttäytyminen ja sen estäminen ruokin- takokeissa, joissa käytetään Insentec RIC-laitteistoa. Maataloustieteen päivät 2014. Helsinki, Suomi, 09.01.2014.
Hämäläinen, W.: Globaali optimonti haastaa lokaalit menetelmät: luotettavampia riippuvuuksia tehosta tinkimättä. Tietojenkäsittelytiede 34 (huhtikuu): 6-16, 2012.
Hämäläinen, W.: Kokemuksia vaihto-opettajana Namibiassa. Sanansaattaja, Joensuun yliopisto, 2005.
Hämäläinen, W.: Miksi tietojenkäsittelytieteessä on niin vähän naisia? Tietojenkäsittelytieteen päivät 2004, International proceedings series 5, University of Joensuu, Department of Computer Science, s. 64-68.
Hämäläinen, W.: Miksi matematiikkaa tietojenkäsittelytieteessä? Raportti tutkinnonuudistajille, 2004.
Source code for efficient search of non-redundant, statistically significant dependency rules (association rules which express statistical dependencies). Notice: The programs do not need any minimum frequency thresholds or restrictions on the rule length, and you can find totally new information in difficult data sets.