Enhanced Reading Comprehension by Automatic Web Searches While Reading
Original Publication Date: 2000-Jul-01
Included in the Prior Art Database: 2003-Jun-18
Disclosed is a method and system for enhancing reading comprehension of text stored in softcopy form. The basic idea is to pre-fetch all possibly relevant information that might help the reader. Authors of documents that are accessed through the Web do not always provide hyperlinks or explanations for every single term they use. Senders of e-mail seldom provide any links. Thus, readers often employ search engines to find on the Web, or on intranets, some hints with regard to the meaning of various terms mentioned in the text they are currently reading. Waiting for the search results can be time consuming and may interfere with the reading process. The method is based on the following elements: 1. Pre-fetching search results. The invention is a system and method that assists the reader of an electronic document (not necessarily on the web; it can be e-mail or a file found on a shared file system) by pre-fetching search results from web searches generated either automatically from the text of the document, or by the reader himself. Thus, the system caches the results on the reader's machine and makes them immediately available when the reader encounters the term. 2. Ranking of terms to be searched. The system pre-analyzes the terms occurring in the text and sorts them according to their likelihood to be understood by the reader. This ranking is based on prior knowledge by the system with respect to the frequency of the occurrence of the term in a reference corpus of documents. In particular, when no specific information about a reference corpus is available, the general frequency in the entire literature is used. In an e-mail application, the system uses the e-mail archive of the person as a reference corpus.