Knowledge discovery about scientific papers or proceedings referenced NASA/DAAC data with a rule-based classifier

摘要

Knowledge discovery from online journals, abstracts and citation indices, cross-referenced with the NASA Distributed Active Archive Center (DAAC) user/order database to close the data-knowledge loop. Knowledge discovery in database (KDD) has been defined as the nontrivial process of discovering valid, novel, potentially useful, and ultimately understandable patterns from data. The core step of the KDD process is data mining. Data mining is all about extracting patterns from an organization's stored or warehoused data. These patterns can be used to gain insight into aspects of the organization's operations and predict outcomes for future situations. Patterns often concern the categories to which situations belong. For example, here is the situation, to decide if a journal paper used the NASA DAAC data or not, starting from the Goddard DAAC user/order database record, a rule-based classifier was developed and rules were found firstly with training samples, then these rules were applied to recognize new patterns.