Machine-Learning
6 posts tagged with "Machine-Learning"
- Aug 11, 2024
👩🏻💻 The Hidden Complexity of Tax Data Analytics: 10 Lessons from the Trenches 👩🏻🏭
The analysis of tax data often yields contradictory paradoxes, requiring both numerical precision as well as analytical interpretation of tax regulations. With Pillar 2 looming, understanding best practices for tax data collection, analysis, calculation, and reporting is paramount. Whether you’re an Excel expert, an Alteryx aficionado, proficient with Python, or skilled using SQL, the following is some hard won practical tips for working with real-world tax data, from data cleaning to building machine learning tax models.
- Apr 7, 2024
Navigating the AI Frontier in Transfer Pricing: a Brief Case Study
Disclaimer: The views and opinions expressed on this blog are strictly my own and do not reflect those of any past or current employer. The content provided here is for informational purposes only and is not intended to represent the strategies, intentions, or opinions of any organization that I am or have been associated with. The use of artificial intelligence (AI) permeates many facets of our lives and work. Here I explore the real-world application of AI in the specialist field of transfer pricing (TP) comparable identification process using KPMG’s recent patent (US 11720842 B2 “System and Method for Identifying omparables”).
- Nov 21, 2023
Digital transformation: how IT can shape the tax profession (🗞️published)
I recently had the honour of contributing an article titled “Digital Transformation: How IT Can Shape the Tax Professional” to the Tax Advisor magazine, published on 20 September 2023. This piece was aimed at demystifying the complex jargon surrounding AI, machine learning, and automation to general tax professinals. In this publication, I delve into how the emerging technologies can revolutionize the tax profession, enhancing accuracy, efficiency, and accessibility. The piece examines a range of technological innovations, from robotic process automation to advanced data analytics and the emergence of Generative AI and Large Language Models.
- Jun 21, 2021
Building a Withholding Tax Determination Engine
I. What is withholding tax? Withholding tax (“WHT”) refers to an amount of tax withheld at source at a certain rate determined by local tax law. The most common example is payroll taxes withheld and paid by employers on employee wages. Withholding taxes also apply where sellers and buyers are situated in different jurisdictions, with rates varying by country or due to specific double tax treaties between tax authorities. WHT can be applied to different types of remuneration, such as dividend, interest, royalty, commission, fees, and rentals, as well as transactions, i.
- Mar 25, 2021
Building an HS Code Classification Model
I. What are HS codes? International trade relies on Harmonized System (“HS”) codes to classify goods transported across borders. Incorrect HS code classification may result in penalties and seizure of goods. HS codes are used by all countries globally to classify imported and exported goods down to 6-digits. The HS code is split into three two-digit groups: First two digits - categorize product (e.g. “61” for Apparel & Clothing) Second two digits - classify further (e.
- Mar 10, 2020
Machine Learning Governance in Tax
Many areas of tax incorporated machine learning, but it is not a silver bullet. What are the potential risks of adopting a machine learning approach to replace manual efforts, and how could we mitigate such risks? This blog post discusses some governance measures I put in place to address these questions. Common applications The most frequent machine learning application I came across in tax falls into the category of classification prediction modelling.