Website content navigation chatbot
Public engagement and open parliament
Italy - Senate
Use case ID: 025
Author: Senate of Italy
Date: 12 June 2024
Objective:
Assist users in navigating the Senate website by providing an artificial intelligence (AI)-powered chatbot that offers content recommendations and guidance based on user queries in natural language.
Actors:
- Senate website users (citizens, researchers and journalists)
- Senate IT and web development team
Prerequisites:
- Existing Senate website with accessible content
- Trained large language model (LLM)-based AI model for understanding and processing natural-language queries
- Database of Senate documents and website content
- Internet accessibility for users
Scenario:
- The user accesses the Senate website.
- User initiates a chat session with the AI-powered chatbot, asking a question or seeking guidance (e.g. “How can I find recent bills on health care?”).
- The LLM-based AI model processes the natural-language query to understand the intent and key terms.
- The AI chatbot provides a relevant response, directing the user to specific sections, documents or links on the website.
- If the user’s request is complex, the chatbot may ask follow-up questions to narrow down the results.
- The chatbot guides the user step by step to the desired content or provides direct links.
- The user can continue to ask additional questions or seek further assistance within the same chat session.
- The system logs the interaction for continuous improvement of the chatbot.
Alternate flows:
- If the LLM-based AI model cannot understand the query, it prompts the user to rephrase or provides suggestions.
- If the user requests information not available on the website, the chatbot can suggest alternative sources or direct them to contact support.
Expected results:
- User satisfaction is improved owing to quick and accurate content navigation assistance.
- Engagement and interaction with the Senate website are increased.
- It takes less time and effort for users to find specific information.
- Accessibility for users unfamiliar with the website layout or content structure is improved.
Potential challenges:
- Ensuring the LLM-based AI model can accurately understand diverse phrasing and terminologies
- Handling ambiguous queries or those with multiple possible interpretations
Data requirements:
- Historical user interactions and queries for training and improving the LLM-based AI model
- Database of Senate documents, laws and other relevant information.
Integrations with other systems:
- Existing Senate website infrastructure
- LLM-based AI processing systems and models
- User interface components for the chatbot
Success metrics:
- Query response time
- User satisfaction ratings and feedback
- Accuracy and relevance of chatbot responses
- Increase in the number of users engaging with the chatbot
- Reduction in user queries requiring manual intervention or support