Every ticket, to the right team.
TicketSense uses hierarchical attention and deep neural networks to read whole IT tickets, predict what the user needs and route each one to the right team instantly, explaining every decision.
The problem
Help desks drown in mixed-format tickets.
Public-sector IT help desks receive large volumes of tickets that mix free text, structured fields and attachments. Staff sort them by hand, so many land with the wrong team and take longer to resolve.
classification accuracy on mixed-format IT tickets.
fewer misclassifications than existing approaches.
training cycle time, fast enough for frequent retraining.
Source: the founding research paper (IEEE ICCR, 2025).
How it works
How it works
A clear path from data to decision.
Ingest
Tickets arrive with text, data fields and attachments.
Read
Hierarchical attention reads the whole ticket across all formats.
Classify
Deep neural networks predict the category and the user's need.
Route
The ticket goes to the right team instantly, with an explanation.
Product
What makes it different.
Every feature is grounded in the founding research.
Multimodal understanding
Handles text, structured fields and attachments together.
High accuracy
Up to 97.1% accuracy and an F1-score of 0.94.
Instant routing
Sends each ticket straight to the right resolver team.
Explainable
Shows why each ticket was classified the way it was.
Who it's for
Who it's for.
Government IT help desks
Faster triage for public-service IT requests.
Enterprise service desks
Less manual sorting and fewer re-assignments.
Managed service providers
Scale ticket handling across many clients.
The research
Built on peer-reviewed research.
- Automated multimodal ticket classification for public IT service management (ITSM).
- Uses hierarchical attention and deep neural networks on mixed-format inputs.
- Reached up to 97.1% accuracy and an F1-score of 0.94.
- Cut misclassifications by more than 25% with training cycles under 2.5 seconds.
Sallibi, A. D., et al. (2025). Automated Multimodal Ticket Classification in Public ITSM Systems Using Hierarchical Attention and Deep Neural Networks on Mixed-Format Inputs. IEEE International Conference on Computing and Robotics (ICCR) 2025.
Team
The team behind it.
Founded in 2025 by a student team from the faculty below.
Faculty of Science (Computer Science)
Deep learning, natural language processing and ITSM integration.
FAQ
Questions we hear often.
Does it work with our ticketing system?
It is designed to plug into common ITSM platforms through their APIs.
Can agents see why a ticket was routed?
Yes. Every decision comes with an explanation.
Is this a finished commercial product?
It is a research-based student startup moving from validated prototype toward pilots with real users.
Can we partner or run a pilot?
Yes. We are looking for organisations willing to test the solution in real conditions and share feedback.
Bring TicketSense to your organisation.
We are onboarding pilot partners now. Read the research and get in touch.