CROWD-SOURCED TEXT ANNOTATION SYSTEM FOR USE BY INFORMATION EXTRACTION APPLICATIONS

摘要

The vocabulary of pertinent terms used to highlight/filter medical records in a text annotation system is continually updated based on user feedback. To maximize the effectiveness of this updating, feedback is extracted from all users of the system, thereby providing a group-sourced vocabulary of pertinent terms. As each user modifies the provided vocabulary of pertinent terms to customize the text annotation system to conform to the users preferences, the modifications are collected and communicated to the provider of the vocabulary of pertinent terms. The provider of the vocabulary of pertinent terms assimilates the modifications implemented by each user of the word annotation system to determine whether to modify the vocabulary of pertinent terms for subsequent users of the common vocabulary of pertinent terms.