BioMining: Data Mining for Biomedical Problems  Page description

Help  Print 
Back »

 

Details of project

 
Identifier
111710
Type PD
Principal investigator Buza, Krisztián Antal
Title in Hungarian BioMining: Gépi tanulás orvosbiológiai feladatokra
Title in English BioMining: Data Mining for Biomedical Problems
Keywords in Hungarian gépi tanulás, osztályozás, link prediction, mátrix kitöltés, orvosbiológiai adatok, idősorok, electroencephalográf (EEG), electrokardiográf (EKG), gene expression data
Keywords in English machine learning, classification, link prediction, matrix completion, biomedical data, time series, electroencephalograph (EEG), electrocardiograph (ECG), gene expression data
Discipline
Information Technology (Council of Physical Sciences)100 %
Ortelius classification: Informatics
Panel Informatics and Electrical Engineering
Department or equivalent Brain Imaging Centre (Research Center of Natural Sciences, Hungarian Academy of Sciences)
Starting date 2014-09-01
Closing date 2017-08-31
Funding (in million HUF) 15.434
FTE (full time equivalent) 1.50
state running project





 

Final report

 
Results in Hungarian
A BioMining projektben adatbányászati eljárások orvosbiológiai alkalmazásaival foglalkoztunk. A csomósodás-alapú gépi tanulás paradigmáját követve fejlesztettünk új eljárásokat orvosbiológiai feladatokra, beleértve a génkifejezés-adatok és orvosbiológiai idősorok osztályozását, valamint a hatóanyagok és farmakológiai támadáspontok közötti kapcsolatok statisztikai predikcióját. A projekt során vizsgált osztályozási technikák a diagnosztikai eljárásokat támogathatják, míg a hatóanyagok és farmakológiai támadáspontok közötti kapcsolatok predikciója a gyógyszerfejlesztés folyamatát segítheti. A kutatási tervben vállalt összes feladatot megfelelően teljesítettük. Kiemeljük továbbá, hogy: (i) 9 impakt faktoros (Web of Science által indexált) folyóiratcikket jelentettünk meg, többek között a Knowledge-Based Systems-ben, Neurocomputing-ban, ill. a Frontiers in Neuroscience-ben, és további 8 konferencián/workshop-on mutattuk be munkánkat, (ii) 16 rövid videót vettünk fel, amelyek egy része egy online előadást alkot a csomósodás-alapú gépi tanulásról (http://www.biointelligence.hu/course.html), más része pedig a projekt legfontosabb eredményeit mutatja be (https://www.youtube.com/playlist?list=PLNWnqkAEYZk1ENQAcydQMHdgQ_KV36-oU); (iii) kifejlesztettük a PyHubs szoftverkönyvtárat (http://www.biointelligence.hu/pyhubs/) és további szoftverkódokat publikáltunk.
Results in English
The BioMining project focused on data mining for biomedical tasks. In particular, we envisioned to develop new hubness-aware machine learning techniques for biomedical tasks, including the classification of gene expression data and biomedical signals as well as drug-target interaction prediction. Classification is related to medical diagnosis, whereas drug-target interaction prediction may increase the efficiency of drug development by delivering promising hypothesises. All tasks of the research plan were implemented appropriately. Regarding the outcome of the project, we point out that: (i) 9 articles have been published in journals indexed by Web of Science, including Knowledge-Based Systems, Neurocomputing and Frontiers in Neuroscience, and further 8 conference or workshop papers were presented at international conferences or workshops; (ii) we recorded 16 short videos with a total length of approx. 100 minutes: these videos are organized into two playlists: one of them is an online lecture about hubness-aware machine learning (http://www.biointelligence.hu/course.html), while the other one summarizes key achievements of the project (https://www.youtube.com/playlist?list=PLNWnqkAEYZk1ENQAcydQMHdgQ_KV36-oU); (iii) we developed the PyHubs software library (http://www.biointelligence.hu/pyhubs/) and published further software codes.
Full text https://www.otka-palyazat.hu/download.php?type=zarobeszamolo&projektid=111710
Decision
Yes





 

List of publications

 
Krisztian Buza, Ladislav Peska: Drug–target interaction prediction with Bipartite Local Models and hubness-aware regression, Neurocomputing, Volume 260, 284-293, 2017
Krisztian Buza: Classification of Gene Expression Data: A Hubness-aware Semi-Supervised Approach, Computer Methods and Programs in Biomedicine, Volume 127, 105-113, 2016
Ladislav Peska, Krisztian Buza, Julia Koller: Drug-Target Interaction Prediction: a Bayesian Ranking Approach, Computer Methods and Programs in Biomedicine, Vol. 152, pp. 15-21, 2017
Regina J. Meszlényi, Petra Hermann, Krisztian Buza, Viktor Gál, Zoltán Vidnyánszky: Resting State fMRI Functional Connectivity Analysis Using Dynamic Time Warping, Frontiers in Neuroscience, Volume 11, Article 75, 2017
Nenad Tomasev, Krisztian Buza, Dunja Mladenic: Correcting the Hub Occurrence Prediction Bias in Many Dimensions, Computer Science and Information Systems, Vol. 13, Issue 1, 2016
Krisztian Buza, Júlia Koller: Classification of Electroencephalograph Data: A Hubness-aware Approach, Acta Polytechnica Hungarica, Vol. 13, No. 2, pp. 27-46, 2016
Krisztian Buza, Noémi Ágnes Varga: ParkinsoNET: Estimation of UPDRS Score using Hubness-aware Feed-Forward Neural Networks, Applied Artificial Intelligence, Volume 30, Issue 6, pp. 541-555, 2016
Krisztian Buza, Alexandros Nanopoulos, Gabor Nagy: Nearest neighbor regression in the presence of bad hubs, Knowledge-Based Systems, Volume 86, 250-260, http://www.sciencedirect.com/science/article/pii/S0950705115002282, 2015
Nenad Tomasev, Krisztian Buza: Hubness-aware kNN classification of high-dimensional data in presence of label noise, Neurocomputing, Volume 160, 157-172, http://www.sciencedirect.com/science/article/pii/S0925231215001228, 2015
Krisztian Buza, Julia Koller, Kristof Marussy: PROCESS: Projection-Based Classification of Electroencephalograph Signals, Artificial Intelligence and Soft Computing, Lecture Notes in Computer Science, Vol. 9120, pp. 91-100, Springer, http://link.springer.com/chapter/10.1007/978-3-319-19369-4, 2015
Krisztian Buza: Semi-supervised Naive Hubness-Bayesian k-Nearest Neighbor for Gene Expression Data, to appear in the Proceedings of the 9th International Conference on Computer Recognition Systems (CORES), Springer, 2015
Krisztian Buza, Noémi Ágnes Varga: Machine Learning for the Estimation of UPDRS score, VII. Dubrovnik Conference on Cognitive Science (DUCOG), 2015
Krisztian Buza: Hubness: An Interesting Property of Nearest Neighbor Graphs and its Impact on Classification, 9th Japanese-Hungarian Symposium on Discrete Mathematics and Its Applications, invited talk, 2015
Kristof Marussy, Ladislav Peška, Krisztian Buza: Recommendations of Unique Items Based on Bipartite Graphs, 9th Japanese-Hungarian Symposium on Discrete Mathematics and Its Applications, 2015
Krisztian Buza, Kristof Marussy: PROGRESS: Projection-Based Gene Expression Classification, Innovations in Medicine Conference, 2014
Krisztian Buza, Alexandros Nanopoulos, Gabor Nagy: Nearest neighbor regression in the presence of bad hubs, Knowledge-Based Systems, Volume 86, 250-260, http://www.sciencedirect.com/science/article/pii/S0950705115002282, 2015
Nenad Tomasev, Krisztian Buza: Hubness-aware kNN classification of high-dimensional data in presence of label noise, Neurocomputing, Volume 160, 157-172, http://www.sciencedirect.com/science/article/pii/S0925231215001228, 2015
Krisztian Buza, Julia Koller, Kristof Marussy: PROCESS: Projection-Based Classification of Electroencephalograph Signals, Artificial Intelligence and Soft Computing, Lecture Notes in Computer Science, Vol. 9120, pp. 91-100, Springer, http://link.springer.com/chapter/10.1007/978-3-319-19369-4, 2015
Krisztian Buza: Semi-supervised Naive Hubness-Bayesian k-Nearest Neighbor for Gene Expression Data, to appear in the Proceedings of the 9th International Conference on Computer Recognition Systems (CORES), Springer, 2015
Krisztian Buza: Classification of Gene Expression Data: A Hubness-aware Semi-Supervised Approach, Computer Methods and Programs in Biomedicine, Volume 127, 105-113, 2016
Krisztian Buza, Júlia Koller: Classification of Electroencephalograph Data: A Hubness-aware Approach, Acta Polytechnica Hungarica, Vol. 13, No. 2, 2016
Nenad Tomasev, Krisztian Buza, Dunja Mladenic: Correcting the Hub Occurrence Prediction Bias in Many Dimensions, Computer Science and Information Systems, Vol. 13, Issue 1, 2016
Krisztian Buza, Noémi Ágnes Varga: ParkinsoNET: Estimation of UPDRS Score using Hubness-aware Feed-Forward Neural Networks, Applied Artificial Intelligence, Volume 30, Issue 6, 2016
Krisztian Buza: Drug-Target Interaction Prediction with Hubness-aware Machine Learning, 11th IEEE International Symposium on Applied Computational Intelligence and Informatics, 2016
Krisztian Buza, Dora Neubrandt: How You Type Is Who You Are, 11th IEEE International Symposium on Applied Computational Intelligence and Informatics, 2016
Krisztian Buza: Person Identification Based on Keystroke Dynamics: Demo and Open Challenge, Forum at the 28th International Conference on Advanced Information Systems Engineering (CAiSE'16), 2016
Regina Meszlényi, Ladislav Peska, Viktor Gal, Zoltán Vidnyánszky, Krisztian Buza: A model for classification based on the functional connectivity pattern dynamics of the brain, Proceedings of the Third European Network Intelligence Conference, 2016
Regina Meszlényi, Ladislav Peska, Viktor Gal, Zoltán Vidnyánszky, Krisztian Buza: Classification of fMRI data using Dynamic Time Warping based functional connectivity analysis, Proceedings of the 24th European Signal Processing Conference, 2016
Krisztian Buza: Semi-supervised Naive Hubness-Bayesian k-Nearest Neighbor for Gene Expression Data, Proceedings of the 9th International Conference on Computer Recognition Systems CORES 2015, pp. 101-110, Springer, 2015
Krisztian Buza, Dora Neubrandt: How You Type Is Who You Are, 11th IEEE International Symposium on Applied Computational Intelligence and Informatics, pp. 453-456, 2016
Rodica Ioana Lung, Mihai Suciu, Regina Meszlényi, Krisztian Buza, Noémi Gaskó: Community structure detection for the functional connectivity networks of the brain, Parallel Problem Solving from Nature - PPSN XIV, pp 633-643, Springer, 2016
K. Buza, D. Neubrandt: A New Proposal for Person Identification Based on the Dynamics of Typing: Preliminary Results, Theoretical and Applied Informatics, Vol. 28, No. 1-2, 2016




Back »