Test datasets for indexing of Chinese medical articles by 4 large models

Four large models were used to extract and summarize the key information of 100 randomly selected Chinese medical research articles, and the accuracy of extraction and summary was scored.The dataset included:Basic information of 100 articles;Prompts constructed based on the abstracts and research methods of 100 literatures;JSON data containing indexing information generated from 4 large models;Key information extracted from JSON data and AI summary text;Scoring of the indexing results of each model.Four LLM are ChatGML4.0, ChatGPT4o, iFLYTEK Spark, Kimi. Four large models were used to extract and summarize the key information of 100 randomly selected Chinese medical research articles, and the accuracy of extraction and summary was scored.The dataset included:Basic information of 100 articles;Prompts constructed based on the abstracts and research methods of 100 literatures;JSON data containing indexing information generated from 4 large models;Key information extracted from JSON data and AI summary text;Scori