Albert Cheng

What can be legally used to physically tracking you?

Am I protected from prying eyes?

In this blog, I explore the main data used for tracking individuals and whether it is actually legal in Australia to do this.

Posted by Albert Cheng on 21 October 2024

Target Identified! How Advanced Fingerprinting is Used to Identify Cyber Attackers

Identifying the bad actors!

In this blog I discuss about how fingerprinting is used to counter against cyber attacks.

Posted by Albert Cheng on 17 March 2023

Unleashing the Power of AI - Learning to Learn with Little Real-World Examples

Overcoming real-world limitations of data availability

I wanted to touch upon this topic, as it is a very real-world challenge experienced by most ML projects and systems.

Posted by Albert Cheng on 13 July 2022

The Machine Learning Minefield - How to Avoid Getting Hit by Machine Learning Poisoning

Data corruption that destroys machine learning systems

In this blog post, I explore the concept of machine learning poisoning and how it is a very dangerous form of attack!

Posted by Albert Cheng on 22 March 2022

The Art of Keeping Secrets - Data Anonymization and Synthetic Data Explained

How to safely handle personally identifiable information

In this blog post, I go over techniques in anonymising data and creating synthetic data to comply with data protection laws

Posted by Albert Cheng on 07 February 2022

How the Pandemic Impacted Human Traffic

Diving into mobility traffic and seeing how COVID-19 impacted human movement

In this blog post, I explore how city traffic changed in light of the COVID-19 pandemic.

Posted by Albert Cheng on 27 January 2022

X-Ray Vision for Machine Learning - How Model Interpretability demystifies the ML black-box

The 'why' is just as important as the 'what'

In this blog post, I will explore why explainable AI is important, as well as some techniques in making machine learning models more explainable

Posted by Albert Cheng on 09 September 2021

A guide on Airflow best practices

Go with the flow!

In this blog post, I cover some best practices to get the most out of Airflow, a popular open-source data orchestration system!

Posted by Albert Cheng on 21 May 2021

Special Sauce of ML - Heuristics and Feature Engineering

The best way to improve ML isn't always in the algorithm

In this blog post, I explore the power of hybrid machine learning systems, which combine the benefits of rules-based models and ML model, as well as how great feature engineering can make or break a system.

Posted by Albert Cheng on 18 October 2020

From Chaos to Control - Using Automated Testing for Big Data Pipelines

Like seat belts, once you have em', you appreciate them a lot!

I explore automated testing in the context of big data infrastructure. Both on data and on code!

Posted by Albert Cheng on 30 September 2020

DataOps Part 2 - using data to find data

Data-driven techniques apply to finding data as well.

In part 2 of DataOps, I discuss how data-driven techniques can help you find and identify your data.

Posted by Albert Cheng on 01 August 2020

DataOps - using data to manage data

The way to manage big data is using data

In this blog I explore the concept of DataOps and how you can use data to manage big data infrastructure.

Posted by Albert Cheng on 25 July 2020

Data science/analytics pitfalls Part 2 - a focus on machine learning

This time focusing on machine learning!

In part 2 of my series on common data science/analytics pitfall, we explore some more ML-specific pitfalls.

Posted by Albert Cheng on 15 June 2020

Exploring some common data science/analytics pitfalls

It's a trap!

In data science/analytics, sometimes there are pitfalls that will lead you to the wrong conclusion (or even worse, misleading conclusions)! This blog post will explore some of the more common pitfalls/traps.

Posted by Albert Cheng on 10 June 2020

From Zero to Kimball - A Guide and Tips on Good Data Warehousing Practices

Exploring Kimball dimension modelling best practices

In this article, I discuss how Kimball dimension modelling best practices can really improve your data warehouse design. A well-designed warehouse is a work of art!

Posted by Albert Cheng on 27 May 2020