Sevvandi Kandanaarachchi

Senior Research Scientist

Data61, CSIRO

Hello! Thanks for visiting my site! 👋

I use statistics, mathematics and machine learning to find unusual patterns in data. They are referred to by different names. They can be called anomalies, outliers or novelties. Sometimes they are called events, especially if an unusual activity is happening over time. What are some examples? Well, intrusions (attacks) in computer networks are anomalies, which are sometimes called anomalous events. Credit card fraud is another example. Fraudulent credit card transactions are anomalies when we consider billions of legitimate transactions. Or a malfunctioning sensor can give out an unusual pattern indicating that the sensor needs replacing or that the batteries have run out. A sudden increase in atmospheric aerosols captured by sensors may indicate a bushfire.

Why is it important to find these unusual patterns? If we detect them quickly, we can act upon it. Yes, early detection is really important. Can we detect an intrusion while it is happening? In this case we only have partial information, which is a challenge. Another challenge is that the data can be high dimensional making it difficult to find anomalies. These are some of the research challenges that I work on. I also like working on real world problems, especially ones that are motivated by industry. From 2016 to 2019, I worked with an industry partner on intrusion detection.

I am an applied mathematician. I came to statistical learning from a mathematics background. My PhD was in mean curvature flow, which has a lot of geometry and differential equations involved. I bring my geometric intuition to my current work.

More details can be found on my CV.


  • Anomaly Detection
  • Event detection
  • Streaming Data
  • Meta-learning
  • Dimension Reduction
  • Data Science


  • Graduate Certificate in Data Mining and Applications, 2015

    Stanford University

  • PhD in Mathematics, 2011

    Monash University

  • MSc Preliminary in Mathematics, 2007

    Monash University

  • BSc Eng. in Computer Science and Engineering, 2002

    University of Moratuwa

Research themes

Anomalies and Algorithms

New algorithms for anomaly detection

Cyber Security

Detecting intrusions in computer networks

Sensor data

Are the sensors telling us something different?


Dimension reduction for outlier detection using DOBIN

Instance Space Analysis for Unsupervised Outlier Detection

Recent & Upcoming Talks

Anomalous Networks

Four, fast geostatistical methods - a comparison

Comparison of geostatistical methods for spatial data

From ensembles to computer networks



Tools for spatio-temporal data exploration - an R package.


Network anomaly detection - an R package.


A collection of ensembles for outlier detection - an R package.


Leave-one-out kernel density estimates for outlier detection - an R package.


Outliers in univariate, multivariate and compositional time series - an R package.


Algorithmic IRT - an R package.


Event detection and early classification for streaming data - an R package.


Dimension reduction for outlier detection - an R package.

Blog posts

Anomaly Detection Ensembles

What is an anomaly detection ensemble? It is a bunch of anomaly detection methods put together to get a final anomaly score/prediction. …

Testing an Outlier Detection Method

Suppose you have developed an outlier detection method. What are the ways to test it? You can generate some random data and add a …

Anomaly detection data repositories

In this post we will look at data repositories available for anomaly detection. So, can you use a standard classification dataset for …

Which nonlinear dimension reduction methods preserve outliers inside a sphere?

In this post we will look at how well nonlinear dimension reduction techniques preserve outliers that are placed inside a sphere, when …

Anomaly detection dilemmas

Finding anomalies/outliers in data is a task that is increasingly getting more attention mainly due to the variety of applications …


Optimisation - The Intelligence Behind AI

After a hiatus, I got a chance to do an episode for the Random Sample. I really enjoyed talking to Prof Peter Stuckey. He has done a …

Forecasting Impact podcasts

Recently, I got a chance to co-conduct episodes for the Forecasting Impact podcast with with Mahdi Abolghasemi. We had great fun …

Di Cook's podcast on data visualization and reproducibility

In this episode I spoke to Prof Di Cook about data visualization and reproducibility in research. Di spoke about the challenges in …

Kate Smith-Miles' podcast on OPTIMA and industry

In this episode I spoke to Prof Kate Smith-Miles about OPTIMA. OPTIMA is an ARC Training Centre and it stands for Optimisation …

A chat with Cheryl Praeger

In this episode I spoke to Prof Cheryl Praeger about many things. She spoke about her love for mathematics and told us a bit about her …

Asha Rao's and Sophie Calabretto's two-part podcast on Mathematics Education and Imposter Syndrome

In this two-part podcast I interviewed Prof Asha Rao from RMIT University and Dr Sophie Calabretto from Macquarie University on …

Galit Shmueli's podcast on tech giants hacking our brains

Tim Macuga from ACEMS and I talked to Prof Galit Shmueli from National Tsing Hua University, Taiwan via zoom and made a podcast. This …

Gael Martin's podcast on Bayesian Statistics

During the 2020 working-from-home period Tim Macuga from ACEMS and I interviewed Prof Gael Martin from Monash University via zoom and …

Rob Hyndman's podcast on forecasting

Are you interested in forecasting? Well, Anthony Mays and myself did a podcast with Prof Rob Hyndman from Monash University on this …

Patricia Menéndez chats about Antarctica

Before the 2020 self-isolation period (also known as the lockdown) Tim Macuga from ACEMS and I informally interviewed Dr Patricia …

Space junk podcast

We humans leave such a lot of junk in space. If you’re interested in space junk, you can listen to two podcasts that I …