{"id":20807587,"url":"https://github.com/vdoninav/hse_criminal_cases","last_synced_at":"2026-04-21T12:31:25.597Z","repository":{"id":217754210,"uuid":"744733699","full_name":"vdoninav/hse_criminal_cases","owner":"vdoninav","description":"HSE project on Criminal Cases Investigation using Named Entity Recognition (NER)","archived":false,"fork":false,"pushed_at":"2024-04-12T14:54:33.000Z","size":85977,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2024-04-14T02:19:28.963Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/vdoninav.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2024-01-17T22:33:17.000Z","updated_at":"2024-04-16T05:35:27.753Z","dependencies_parsed_at":"2024-04-16T05:46:05.317Z","dependency_job_id":null,"html_url":"https://github.com/vdoninav/hse_criminal_cases","commit_stats":null,"previous_names":["vdoninav/hse_criminal_cases"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/vdoninav/hse_criminal_cases","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vdoninav%2Fhse_criminal_cases","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vdoninav%2Fhse_criminal_cases/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vdoninav%2Fhse_criminal_cases/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vdoninav%2Fhse_criminal_cases/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/vdoninav","download_url":"https://codeload.github.com/vdoninav/hse_criminal_cases/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vdoninav%2Fhse_criminal_cases/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28002265,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-12-24T02:00:07.193Z","response_time":83,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-17T19:39:02.626Z","updated_at":"2025-12-24T12:16:46.426Z","avatar_url":"https://github.com/vdoninav.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# HSE Criminal Cases Investigation\n\n**by Lebedyuk Eva, Vdonin Aleksei**  \nFaculty of Computer Science, HSE University  \n\n## Coursework 2024  \nThis coursework focuses on the development of a system for **Criminal Cases Investigation** using **Named Entity Recognition (NER)** to extract key entities from judicial texts.\n\n### Overview  \nThe project aims to automate the analysis of judicial documents, specifically focusing on **Article 105, Part 1 of the Russian Criminal Code (\"Murder\")**. It includes the following key components:\n1. **NER Model**: Fine-tuned a BERT-based model for extracting entities such as individuals, legal entities, crimes, laws, and penalties.\n2. **Summarization Module**: Implemented extractive summarization for condensing large judicial texts by more than 90% while preserving key information.\n3. **Web Application**: Developed an interactive interface using **Streamlit** for entity extraction and summarization, enabling efficient document analysis.\n\n### Key Features  \n- **NER Pipeline**: Automatically identifies and categorizes entities in legal texts.\n- **Text Summarization**: Provides concise summaries of judicial documents for faster comprehension.\n- **Interactive Interface**: Allows users to input texts, visualize predictions, and interact with summarized outputs.\n- **Optimized Deployment**: Application optimized for CPU usage, ensuring accessibility on standard systems.\n\n### Contributions  \n- **Lebedyuk Eva**: Led the development of the NER model, including data preprocessing, training, and evaluation.\n- **Vdonin Aleksei**: Developed the web application interface and implemented the summarization module.\n\n### Project Details  \n- **Dataset**: The model was trained on the **RuLegalNER** dataset, adapted for extracting legal entities in Russian texts.\n- **Tech Stack**:\n  - Python (Hugging Face Transformers, Streamlit)\n  - Libraries: BeautifulSoup, PyMyStem, JSON\n  - Infrastructure: Google Colab, Streamlit Cloud\n\n### Challenges  \n- High computational demands for model training.\n- Integration of multiple components into a seamless application.\n\n### Future Work  \n- Enhance entity recall for legal entities and penalties.\n- Expand summarization functionality with hybrid models.\n- Integrate additional entity types, such as dates and participant relationships.\n\n### Link to the Project Web App:  \n[HSE Criminal Cases Web App](https://hse-criminal-cases.streamlit.app)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvdoninav%2Fhse_criminal_cases","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvdoninav%2Fhse_criminal_cases","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvdoninav%2Fhse_criminal_cases/lists"}