Adaptive Information Monitoring and Extraction

描述（由申请人提供）： 现在人们普遍认识到，非常需要更强大的自动化方法来帮助生物医学科学家从科学文献中过滤，查询和提取信息。基于我们过去在生物医学文本挖掘方面的研究成果，我们计划开发新的算法和软件系统，以显着提高生物医学研究人员利用科学文献的能力。特别是，我们计划开发，评估和现场系统，（1）通过从文本源提取和组织信息来帮助注释高通量实验，（2）通过识别相关文章并预测特定查询基因和蛋白质的适当本体代码来帮助基因组数据库管理员。为了支持这些系统，我们计划开发新的基于机器学习的文本挖掘算法，用于训练粗糙标记的数据，并归纳以自然语言表达的特定类型实体之间的关系模型。

DESCRIPTION (provided by applicant): It is now widely recognized that there is a great need for more powerful automated methods to assist biomedical scientists in filtering, querying, and extracting information from the scientific literature. Building on our past research accomplishments in biomedical text mining, we plan to develop new algorithms and software systems that will significantly improve the ability of biomedical researchers to exploit the scientific literature. In particular, we plan to develop, evaluate and field systems that (1) aid in annotating high-throughput experiments by extracting and organizing information from text sources, and (2) assist genome database curators by identifying relevant articles and predicting appropriate ontology codes for specific query genes and proteins. In support of these systems, we plan to develop novel machine-learning based text-mining algorithms for training on coarsely labeled data, and inducing models of relationships among specific types of entities expressed in natural language.