01
Past
MSc Data Science & Analytics at the University of Westminster (2023–2024). Dissertation in NLP and sentiment analysis on Reddit data.
Research & Education
From enterprise software to decision support under uncertainty.
An academic trajectory grounded in enterprise practice — postgraduate work in data science and NLP, current interests in operating-model intelligence, and a future direction toward AI-driven decision support for industrial systems.
Research Trajectory
02
Current
Building decision support into my own products: forecasting and replenishment in Procura and HOP.
03
Future
MPhil/PhD in Computer Science at the University of East London, from January 2027.
Education
MSc, Data Science and Analytics
University of Westminster, London · Aug 2023 — Oct 2024
Postgraduate study in statistical learning, data engineering, natural language processing, and applied analytics — completed alongside concurrent practitioner roles in the United Kingdom.
BSc Computer Software Technology Engineering, University of Applied Science and Technology, 2009–2012
2009–2012
Final Project: User Support System — developed using C# and SQL Server to manage and track user requests within an organisational environment.
Doctoral research (from January 2027)
Offer: MPhil/PhD in Computer Science, University of East London, starting January 2027. Proposed research: an uncertainty-aware digital twin for inventory replenishment in enterprise software, covering dynamic reorder points, risk-adjusted order quantities and adaptive safety stock under demand and lead-time uncertainty. Developed with Dr Seyed Ali Ghorashi. I will test the findings in Procura and HOP. The university has also invited me to teach.
MSc Dissertation
Reddit-based sentiment analysis of the 2024 U.S. Presidential Election and its impact on Middle East tensions.
This dissertation examined how public online discussion responded to the 2024 U.S. Presidential Election, and how that discussion tracked alongside developments in Middle East tensions. I collected discourse from the Reddit API and built an end-to-end NLP pipeline — cleaning the text, scoring sentiment with VADER, and extracting features with TF-IDF across unigram, bigram, and trigram representations. I then compared several machine-learning classifiers to detect shifts in tone, topic, and polarity across the election period, with Random Forest and Support Vector Machine among the strongest performers. Alongside the classification work, I tracked sentiment distribution and discussion volume over time to see how the conversation moved as events unfolded.
Methods: Reddit API data collection · NLP · VADER sentiment scoring · TF-IDF · unigram/bigram/trigram features · ML classification (Random Forest, SVM).
Publication
Conference Paper
A Conceptual Model for Intelligent Knowledge Management in the Big Data Era: Synergizing Artificial Intelligence and Business Analytics.
Philip Worrall · Mohammad Mehrabani · Hadi Shafiee Bafti
Presented at the Second International Conference on Advances in Artificial Intelligence in Engineering and Humanities, Melbourne, Australia, August 2026. Published in Civilica (document No. 2708420).
A conceptual framework that examines how Artificial Intelligence and Business Analytics can be integrated within intelligent knowledge management systems to support decision-making, knowledge sharing, and organisational learning in data-intensive environments.
Status — Presented and published
Research Interests
A research trajectory shaped by enterprise practice — open to collaboration with academic groups working in adjacent areas.
AI-Driven Decision Support Systems
Digital Twins for Enterprise Operations
Uncertainty-Aware ERP Systems
Operational Intelligence
Data-Driven Decision Making
Enterprise Knowledge Systems
Future Direction
My current interests sit at the intersection of enterprise systems, operational decision-making, and applied artificial intelligence.
After years designing capabilities for enterprise software, I keep meeting the same problem: organisations hold plenty of data but still struggle to make good decisions under uncertainty.
This observation has led me towards exploring AI-driven decision support, demand sensing, inventory digital twins, uncertainty-aware planning, and operational intelligence within complex organisational environments.
My longer-term objective is to contribute to practical approaches that help organisations move beyond transaction processing and reporting toward more adaptive, evidence-informed, and operationally effective decision-making.
Current Research Direction
Current research interests focus on AI-driven decision support, uncertainty-aware decision support in enterprise software, inventory digital twins, demand sensing, and operational intelligence for complex organisational environments.