How AI Helped a University Choose Better Social Media Posts

By Nuruzzaman Faruqui

·

Universities compete for students on social media, and one badly received post can leave a lasting negative impression. DPMS is an AI system that reads how audiences feel about different parts of a university and helps choose content they will respond to well.

98.66%sentiment classification accuracy
68.75%increase in positive impressions
18%higher conversion rate

The problem

Social media marketing is now mainstream, including for universities. Audience reactions decide whether a campaign succeeds. Choosing the right content is one of the most important decisions, yet it is often based on guesswork.

What the research did

We built the Data-Driven Promotional Management System (DPMS). An optimised bidirectional LSTM network classifies users’ sentiments about different university divisions from their social media responses. That insight guides which content is most likely to get positive reactions.

What we found

DPMS classified sentiment with 98.66% accuracy. Used to guide content, it increased positive impressions by 68.75%, reduced adverse reactions by 31.25% and increased the conversion rate by 18%. It is the first promotional management system designed for universities.

Who benefits

  • University marketing teams
  • Prospective students, who see more relevant content
  • Other organisations using social media to engage the public

Limits and what comes next

The results come from one institutional context. Sentiment analysis can misread sarcasm and cultural nuance, and should inform rather than replace human judgement.

The team

This research was led by Mohamed Emran Hossain (Daffodil International University) with me, Imran Mahmud, Tony Jan (Torrens University Australia), Md Whaiduzzaman and Alistair Barros (QUT).

Frequently asked questions

What is DPMS?

A Data-Driven Promotional Management System that uses AI to guide which content a university posts on social media.

How does it work?

It reads audience sentiment about different university divisions and recommends content likely to be received positively.

What results did it achieve?

68.75% more positive impressions and an 18% higher conversion rate.

About the author

Nuruzzaman Faruqui is an award-winning AI educator and researcher who has taught artificial intelligence and machine learning at university level since 2018. A two-time winner of the Excellence in Teaching Award (2023, 2024) and winner of the Best Researcher Award (2024) at Daffodil International University, he is now a doctoral researcher in AI at Monash University, Australia. He has published 53 papers and six books, and writes these articles to make AI research easy for everyone to understand. His teaching · About Zaman · Get in touch

Keep reading