Konstantinos Sechidis
AstraZeneca Data Science Fellow
University of Manchester



I am a post-doc in the Machine Learning and Optimization Group holding the AstraZeneca/Manchester Data Science Fellowship. My project focuses on statistical and machine learning methods for subgroup analysis in clinical trials. Prior to that, I did my PhD in the School of Computer Science under the supervision of Dr. Gavin Brown. My research interests are in the area of information theoretic feature selection in different learning environments and particularly focusing on medical and health informatics applications.

Areas of interest: Information Theory, Feature Selection, Multi-label learning, Positive-unlabelled learning, Semi-supervised learning, Under-reporting

Recent news
[September 2017] Attended the 15th Annual Pharmaceutical IT Congress and the Artificial Intelligence in Drug Development Congress organized by Oxford Global, London, UK.
[July 2017] New publication: "Simple strategies for semi-supervised feature selection", published in the Machine Learning Journal. Code available in GitHub!!!
[June 2017] Research visit in the Imprecise Probability Group of IDSIA/SUPSI, invited by Giorgio Corani and Marco Zaffalon. An amazing week full of interesting ideas for future collaborations.
[May 2017] New publication: "Algorithmic challenges in Big Data analytics", accepted in ESANN 2017.
[April 2017] New publication: "Dealing with under-reported variables: An information theoretic solution", published in the International Journal of Approximate Reasoning. Code available in GitHub!!!
[March 2017] New publication: "On the Use of Spearman's Rho to Measure the Stability of Feature Rankings", accepted in IbPRIA 2017.
[February 2017] New publication: "Exploring the consequences of distributed feature selection in DNA microarray data", accepted in IJCNN 2017.
[January 2017] New publication: "Disentangling Prognostic and Predictive Biomarkers Through Mutual Information", accepted in Informatics for Health 2017.
[December 2016] New publication: "Ranking Biomarkers Through Mutual Information", accepted in NIPS 2016 Workshop on Machine Learning for Health (ML4HC), Barcelona.
[November 2016] My PhD got the Best Thesis Prize of the School of Computer Science at the University of Manchester 2016 (sponsored by IBM).
[November 2016] Invited research visit in the Intelligent Systems group of AUTH, invited by Grigorios Tsoumakas. I presented my PhD work and exchange some interesting ideas for future collaboration.
[September 2016] Invited research visit in the Imprecise Probability Group of IDSIA, invited by Giorgio Corani. An amazing week full of fruitful exchange of ideas.
[August 2016] I am co-organising the special session Algorithmic Challenges in Big Data Analytics at the European Symposium in Artificial Neural Networks (ESANN 2017).
[July 2016] New publication: "Estimating mutual information in under-reported variables", accepted in PGM 2016.
[January 2016] Start working as a research associate in the EPSRC project AnyScale Apps. Currently, I am collaborating with Matt, Emily and Gavin on exploring information theoretic approaches for correcting under-reporting.
[October 2015] Finished my PhD on Hypothesis Testing and Feature Selection in Semi-Supervised Data!!! You can download it here.
[September 2015] Attended Machine Learning for Personalized Medicine Summer School organized by the University of Sheffield, UK.
[July 2015] Gave an invited talk about testing and estimation in partially labelled data, in Health e-Research Centre, Manchester, UK.
[June 2015] New publication: "Markov blanket discovery in positive-unlabelled and semi-supervised data", accepted in ECML/PKDD 2015. Acceptance rate 89/383 (23.2%).
[May 2015] Gave the tutorial "An introduction to statistical testing and estimation" for Computer Science PhD students of the University of Manchester, details here.
[November 2014] Got the runner-up best paper prize in the School of Computer Science at the University of Manchester 2016 (sponsored by IBM).
[November 2014] Gave an invited talk about positive unlabelled testing, in Xerox Research Centre Europe (XRCE), Meylan, France.
[October 2014] Gave the lecture "Statistical Hypothesis Testing" for the postgraduate course COMP61011: Foundations of Machine Learning of the University of Manchester.
[September 2014] Attended ECML/PKDD 2014 in Nancy, France. Our work got the Best Student Paper Award (sponsored by Springer).
[June 2014] New publication: "Statistical Hypothesis Testing in Positive Unlabelled Data", accepted in ECML/PKDD 2014. Acceptance rate 115/483 (23.8%).
[April 2014] New publication: "Information theoretic feature selection in multi-label data through composite likelihood", accepted in SSPR 2014.
[January 2014] Attended Gaussian Process Winter School in University of Sheffield, UK.
[October 2013] Gave the lecture "Learning From Multi-label Data" for the postgraduate course COMP61011: Foundations of Machine Learning of the University of Manchester.
[June 2012] Attended ICML in Edinburgh, UK.
[April 2012] Attended 19th Machine Learning Summer School in La Palma, Spain.

I would like to thank the one and only Sarah Nogueira for helping me create this webpage :)