Researchers have developed an AI Appeals Processor designed to automate the classification and routing of citizen appeals in government services. This deep learning model, utilizing multilingual BERT, achieved 82% accuracy on a Russian-language dataset, significantly outperforming human operators. While deployed with human oversight, the system reduced end-to-end handling time by over 50%, with most remaining errors attributed to the appeal taxonomy itself rather than the model's performance. AI
IMPACT Automates government appeal classification, reducing handling times and potentially improving efficiency in public services.
RANK_REASON The cluster contains a research paper detailing a deep learning approach to a specific problem. [lever_c_demoted from research: ic=1 ai=1.0]
- AI Appeals Processor
- fastText
- Hugging Face
- long short-term memory
- multilingual-BERT
- support vector machine
- Volodymyr Bezkorovainyi
- Word2vec
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →