Alexander FischPhD student, Associate Lecturer
Born and raised in Belgium, I moved to the UK in 2012 to read Mathematics at Trinity College, Cambridge. Over my undergraduate years, I progressively discovered I had an affinity for statistics and working with data in particular. This prompted me to do data science internships with Atass Sports and LBBW. I then joined STOR-i in 2016 after graduating form Cambridge with my MMath.
My PhD focusses on anomaly detection and is in collaboration with BT. My supervisors are Paul Fearnhead and Idris Eckley.
Personal webpage: http://www.lancaster.ac.uk/~fisch/
My research focusses on anomaly detection. Anomalies are observations which do not conform with the general pattern of a dataset. They can be due to wrong recording or be true outliers. Historically, most datsets were of manageable size, which allowed for anomalies and outliers to be identified manually. Today the low price of sensors and the internet has increased the number and size of datsets making automated anomaly detection necessary. Anomaly detection is of particular importance in many areas including fault detection in mechanical devices such as enginges, as well as fraud detection and prevention in online banking.
anomaly: Detecting Anomalies in Data
Grose, D., Eckley, I., Fearnhead, P., Fisch, A. 21/09/2018
A linear time method for the detection of point and collective anomalies
Fisch, A.T.M., Eckley, I.A., Fearnhead, P. 7/06/2018 In: arXiv.
- Changepoints and Time Series
- STOR-i Centre for Doctoral Training