evidoria

← Back to browse

Good practice Imported

SQUbot — Sultan Qaboos University's Bilingual AI Chatbot for Student Services

Oman · Muscat · See the Oman profile · See the Muscat profile

Evidence: Observational / pre–post Top 81% 33/100 · Ask Evidence Copilot about this practice

Sultan Qaboos University piloted SQUbot, a bilingual Arabic/English chatbot built on IBM Watson NLU, to answer general and OTP-verified personalised student-service queries; the pilot recorded 84.9% intent-recognition accuracy across real student interactions.

84.9 %
Intent-recognition accuracy in pilot testing (pilot testing)
60.8 %
Share of queries for general university services (pilot testing)
39.2 %
Share of queries requesting personalised information (pilot testing)
SQUbot — Sultan Qaboos University's Bilingual AI Chatbot for Student Services

Details

Maturity
Pilot
Promoter
Sultan Qaboos University
Period
2024–2025 (pilot)
Keywords
higher education, student services, conversational AI, NLP

Context

Sultan Qaboos University (SQU) in Muscat, Oman, developed SQUbot to ease pressure on staffed student-service desks handling a high volume of routine queries from a bilingual Arabic- and English-speaking student body.

Objectives

Reduce the burden on staffed student-service desks by automating routine student queries while securely handling personalised information requests.

Activities

The system uses IBM Watson's natural language understanding and deep learning to interpret queries, and distinguishes between general information requests and personalised queries, which are unlocked through one-time-password (OTP) and student-email verification.

Results

In pilot testing, SQUbot achieved 84.9% intent-recognition accuracy; 60.8% of user queries concerned general university services and 39.2% requested personalised information, giving the SQU team a baseline for further development.

Conclusions

The published study is a single-institution pilot: it reports no independent data-protection audit, and the authors describe the results as a foundation for future large-scale deployment rather than proof of scaled impact.

Implementation
Indicative cost
Low (< €50k)
Time to results
Short (< 1 year) — Reported as a pilot study; authors describe results as a foundation for future large-scale deployment.
Staffing & skills
Sultan Qaboos University project/research team, IBM Watson NLU platform (technology provider)

Conditions for success

  • Bilingual (Arabic/English) natural-language understanding tuned to university-specific queries
  • OTP and student-email verification to gate personalised information requests

Common failure modes

  • No independent data-protection audit reported
  • Single-institution pilot with no evidence yet of scaled deployment

Where it fits

Governance type
public university
Scale
single institution pilot
Income level
high-income

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful