Dr Lowri Williams

Cardiff University

  • Research Data Scientist

About Me

A research data scientist with 4 years of experience in academic research and 2 years in industry-based collaborations supporting small-medium businesses with novel data science solutions.

Collaborated with multidisciplinary experts and contributed to research outputs and findings within several reputable academic journals within disciplines such as natural language processing and cybersecurity.

Research areas and interests: natural language processing, data/text mining, sentiment analysis, language resources, machine learning, adversarial machine learning, feature engineering, classification, classification schemes, data science.


Work Experience

Discribe Hub+

2021 - Current

Research Associate - Data Scientist

Cardiff University


Discribe is a community of researchers and professionals, bringing diverse thinking together to help realise the possibility of a secure digital future for all.

Data Innovation Accelerator

2019 - 2021

Research Associate - Data Scientist

Cardiff University


The Data Innovation Accelerator supports small medium businesses to apply data science techniques to produce tangible benefits.

National Corpus of Contemporary Welsh

2017 - 2019

Research Assistant

Cardiff University


CorCenCC is a collection of language samples gathered from real-life communication to support Welsh speakers, learners, and researchers.

Education

Cardiff University

2013 - 2017

Doctor of Philosophy (PhD)


Funded by EPSRC Doctoral Training Partnership.

Cardiff University

2010 - 2013

BSc Information Systems

Volunteering Experience

Frontiers

2021 - Current

Guest Associate Editor


A member of the editorial board of the Social Data Science: Using Social Networks Data for Social Impact journal.

Towards Data Science

2020 - Present

Voluntary Editor


A member of the editorial team for Towards Data Science, a Medium publication sharing data science concepts, ideas, and code.

Academic Literature

Hardening Machine Learning Denial of Service (DoS) Defences Against Adversarial Attacks in IoT Smart Home Networks

E Anthi, L Williams, A Javed, P Burnap. Computers & Security, Volume 108, 2021.

A Three-Tiered Intrusion Detection System for Industrial Control Systems (ICS)

E Anthi, L Williams, P Burnap, K Jones. Journal of Cybersecurity, Volume 7, Issue 1, 2021.

Adversarial Attacks on Machine Learning Cybersecurity Defences in Industrial Control Systems

E Anthi, L Williams, M Rhode, P Burnap, A Wedgbury. Journal of Information Security and Applications, Volume 58, 2021.

Comparing the Utility of Different Classification Schemes for Emotive Language Analysis

L Williams, M Arribas-Ayllon, A Artemiou, I Spasić. Journal of Classification 36 (3), 619-648. 2019.

A Supervised Intrusion Detection System for Smart Home IoT Devices

E Anthi, L Williams, M Słowińska, G Theodorakopoulos, P Burnap. IEEE Internet of Things Journal 6 (5). 2019.

Pulse: An Adaptive Intrusion Detection for the Internet of Things

E Anthi, L Williams, P Burnap. Living in the Internet of Things: Cybersecurity of the IoT-2018 (pp. 1-4). IET. 2018.

Idiom Based Features in Sentiment Analysis: Cutting the Gordian Knot

I Spasić, L Williams, A Buerki. IEEE Transactions on Affective Computing 11 (2), 189-199. 2017.

Pushing the Envelope of Sentiment Analysis Beyond Words and Polarities

L Williams. PhD Thesis. Cardiff University. 2017.

The Role of Idioms in Sentiment Analysis

L Williams, C Bannister, M Arribas-Ayllon, A Preece, I Spasić. Expert Systems with Applications 42 (21), 7375-7385. 2015.


Personal Projects

Portal COVID-19 Cymru

Portal COVID-19 Cymru reports daily Welsh COVID-19 data and statistics from different sources.

Spotify Sentiment Analysis

Spotify Sentiment Analysis applies sentiment analysis to your favourite songs in your Spotify playlist.

Blogs

WordNet: A Lexical Taxonomy of English Words

What is WordNet and why is it useful?

Sentiment Analysis: Aspect-Based Opinion Mining

Investigating sentiment analysis and topic modelling techniques.

Sentiment Analysis: Idioms and their Importance

An investigation into the role of idioms in sentiment analysis.

Adversarial Attacks on SMS Spam Detectors

Applying adversarial learning to target SMS spam detectors.

Coronavirus: Using Machine Learning to Triage COVID-19 Patients

Analysing and classifying real patient cough sounds.

Topic Modelling: Going Beyond Token Outputs

An investigation into how to assign topics with meaningful titles.


Spotify Sentiment Analysis

A sentiment analysis project on Spotify data.