{"id":22903465,"url":"https://github.com/sawyerbutton/medical-rumor-explanation","last_synced_at":"2025-04-01T07:23:27.252Z","repository":{"id":184575874,"uuid":"672137264","full_name":"sawyerbutton/Medical-Rumor-Explanation","owner":"sawyerbutton","description":"Medical Rumor Explanation Pipeline","archived":false,"fork":false,"pushed_at":"2023-07-31T01:28:34.000Z","size":8307,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-02-07T04:40:46.312Z","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":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sawyerbutton.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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,"publiccode":null,"codemeta":null}},"created_at":"2023-07-29T03:41:13.000Z","updated_at":"2023-07-30T05:11:37.000Z","dependencies_parsed_at":null,"dependency_job_id":"7297c9fe-d8a9-40ff-af12-5ffe39c80df8","html_url":"https://github.com/sawyerbutton/Medical-Rumor-Explanation","commit_stats":null,"previous_names":["sawyerbutton/medical-rumor-explanation"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sawyerbutton%2FMedical-Rumor-Explanation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sawyerbutton%2FMedical-Rumor-Explanation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sawyerbutton%2FMedical-Rumor-Explanation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sawyerbutton%2FMedical-Rumor-Explanation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sawyerbutton","download_url":"https://codeload.github.com/sawyerbutton/Medical-Rumor-Explanation/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246598950,"owners_count":20803115,"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","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":"2024-12-14T02:36:52.911Z","updated_at":"2025-04-01T07:23:27.235Z","avatar_url":"https://github.com/sawyerbutton.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Medical Rumor Detection Pipeline\n\n## Background\n\n1. The analyzed Rumor file contains 125 valuable data entries.\n2. Each data entry spans 5 dimensions: Time, Rumor Nature, Rumor Content Tag, Counter-rumor Explanation, Rumor Summary, and Original Rumor Text.\n3. The Huatuo model, fine-tuned for this task, is a model adjusted from the LLaMA model enriched with Chinese medical knowledge. This was accomplished by building a Chinese medical command dataset using a medical knowledge graph and the GPT-3.5 API. The LLaMA was then fine-tuned on these commands, enhancing its performance in medical Q\u0026A scenarios.\n\n### Data analysis\n\n1. Rumor Nature and Rumor Content Tag might be largely irrelevant.\n2. Time could be used for logical validation.\n3. In most cases, the Rumor Summary is a correct abstraction of the Original Rumor Text. To save on tokenization, the Rumor Summary can be used as the context for subsequent fine-tuning steps.\n\n## Purpose\n\n1. The goal is to train a model suitable for analyzing medical rumors. It should not only categorize the rumors but also provide a level of confidence in its judgment (based on a certain threshold).\n\n## Procedure\n\n1. Construct a fine-tuning dataset that meets the requirements of the Huatuo model from the existing data. The current dataset is limited; additional data scraping/creation is required.\n2. Fine-tune the Huatuo model using Lora.\n3. Validate the capabilities of the fine-tuned Huatuo model (accuracy across two dimensions: Counter-rumor Explanation and Rumor Judgment Confidence).\n\n## Todos\n\n- [ ] Scrape medical rumors.\n- [x] Build a dataset for fine-tuning the Huatuo model.\n- [ ] Construct serveral few-shot finetune solutions for medical rumor.\n- [x] Using ChatGLM2-6B model to build a result baseline\n- [ ] Set up the base Huatuo model.\n- [ ] Verify the rumor analysis capabilities and accuracy of the Huatuo model.\n- [ ] Fine-tune the Huatuo model using the dataset.\n- [ ] Validate the capabilities of the fine-tuned Huatuo model.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsawyerbutton%2Fmedical-rumor-explanation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsawyerbutton%2Fmedical-rumor-explanation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsawyerbutton%2Fmedical-rumor-explanation/lists"}