Researchers have developed new decision-theoretic stopping rules for document screening, aiming to optimize search result examination. These policies, derived from the Expected Value of Perfect Information, consider the specific reason for reviewing documents rather than solely focusing on recall targets. Applied to patent examining and systematic reviewing tasks, the new approach demonstrated superior performance over existing methods in experiments using CLEF-IP and medical systematic review datasets, yielding higher net utility. AI
RANK_REASON This is a research paper published on arXiv detailing a new methodology for document screening. [lever_c_demoted from research: ic=1 ai=0.4]
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
