Related Experiment Video
Updated: Apr 6, 2026
05:47
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
Published on: June 13, 2025
1.9K
Clinic expert information extraction based on domain model and block importance model
Yuanpeng Zhang1, Li Wang2, Danmin Qian1
1Department of Medical Informatics, Medical School, Nantong University, 19 Qixiu Road, Nantong 226001, Jiangsu Province, China.
Computers in Biology and Medicine
|August 2, 2015
Summary
This study introduces novel methods for extracting expert clinic information from the Deep Web. A domain model improves form classification, and a block importance model enhances data extraction from response pages.
Area of Science:
- Information Science
- Computer Science
Background:
- Extracting specialized information from the Deep Web presents significant challenges.
- Existing methods struggle with form classification and filtering irrelevant data from search results.
Purpose of the Study:
- To develop and evaluate novel methods for improving the extraction of expert clinic information from the Deep Web.
- To address the challenges of form classification and information extraction from response pages.
Main Methods:
- A domain model, utilizing a tree structure of query interface attributes, was proposed for classifying query interfaces and populating them with domain keywords.
- A block importance model, considering both content and spatial features, was developed to filter noisy information from web pages.
Main Results:
- The proposed domain model achieved a 4.89% higher precision compared to traditional rule-based methods for form classification.
- The block importance model demonstrated a 10.5% higher F1 measure than the XPath method for information extraction.
Conclusions:
- The developed domain model and block importance model significantly enhance the accuracy and efficiency of extracting expert clinic information from the Deep Web.
- These novel approaches offer a robust solution for navigating and retrieving valuable data from complex online sources.
