Classification and preparation for addressing public comments

  • Client typePublic sector
  • IndustryGovernment & Public Services
  • Application areasLegal, Compliance & Risk, Strategy, Planning & Decision-Making
  • AI technologiesGenerative AI, Large Language Models (LLMs), Natural Language Processing (NLP)
  • Business impactsEmployee Enablement & Productivity, Operational Efficiency & Cost Savings
  • Data typesDocuments / Semi-structured Data, Geospatial Data, Textual Data
  • Delivery modelsCustom Development
  • DeploymentsHybrid
  • Key capabilitiesIntelligent Search & Knowledge Retrieval, Recognition, Classification & Tracking
  • Project stagesInitial Production Deployment
  • Solution formsAnalysis, Recommendation, or Report, API / Micro-service Interface, Web Portal / Dashboard

Solution Description

Problem description

The project targeted an agenda that was highly manually intensive for officials. Each submission from the public had to be read, assigned to a relevant category for further processing, and checked for possible similarities or overlaps with previous submissions.

Solution

Design and implementation of an end-to-end solution in the MS Azure environment, aimed at accelerating the processing of incoming submissions from the public and stakeholder authorities. This includes the implementation of semantic classification into 70 categories, duplicate detection, and the semi-automatic preparation of responses. The project also involves implementing an API for data transfer from the on-premise source system, setting up a queue for processing large requests, and developing a web application for user interaction.

Main Users of the Solution

Officials

Project timeframe (months)

6

Additional services

  • Identification and prioritization of suitable use cases
  • Change management and user training
  • Compliance / regulatory support

Implementation

Project Owner on the Client's Side

Head of Functional/Operational Unit

Participation on the Client's Side

  • Business / Product Owner
  • Domain / Process Experts
  • Data & ML specialists
  • Quality, safety, compliance

Form of Supplier Involvement

Complete realization

Impact and Results

Qualitative Benefits

A several-fold acceleration of the process, primarily when handling repetitive, related, or otherwise similar inputs.

Lessons Learned and Recommendations

Key Success Factors

An agile approach to creating a custom solution, accommodating continuous changes to the project scope and individual features.

Recommendation for Others

Focus on selecting a vendor who is willing to collaboratively define the project scope, proactively work towards a solution, and is not afraid to start, even without being able to foresee every possible outcome.

Stay Up to Date with CNAIP. The Czech AI is evolving fast—and we'd love for you to be a part of it. Leave your contact to hear about what's next.

By subscribing, you agree to our Terms of Service.

© cnaip 2025

Want to become a part of Czech AI?

Share your story and showcase what you can achieve with artificial intelligence. Your involvement will inspire others and help us map out the Czech AI scene in its entirety.