tailieunhanh - Báo cáo khoa học: "Matching Readers’ Preferences and Reading Skills with Appropriate Web Texts"

This paper describes Read-X, a system designed to identify text that is appropriate for the reader given his thematic choices and the reading ability associated with his educational background. To our knowledge, Read-X is the first web-based system that performs real-time searches and returns results classified thematically and by reading level within seconds. To facilitate educators or students searching for reading material at specific reading levels, Read-X extracts the text from the html, pdf, doc, or xml format and makes available a text editor for viewing and editing the extracted text. . | Matching Readers Preferences and Reading Skills with Appropriate Web Texts Eleni Miltsakaki University of Pennsylvania Philadelphia . elenimi@ Abstract This paper describes Read-X a system designed to identify text that is appropriate for the reader given his thematic choices and the reading ability associated with his educational background. To our knowledge Read-X is the first web-based system that performs real-time searches and returns results classified thematically and by reading level within seconds. To facilitate educators or students searching for reading material at specific reading levels Read-X extracts the text from the html pdf doc or xml format and makes available a text editor for viewing and editing the extracted text. 1 Introduction The automatic analysis and categorization of web text has witnessed a booming interest due to increased text availability of different formats txt ppt pdf etc content genre and authorship. The web is witnessing an unprecedented explosion in text variability. Texts are contributed by users of varied reading and writing skills as opposed to the earlier days of the Internet when text was mostly published by companies or institutions. The age range of web users has also widened to include very young school and sometimes pre-school aged readers. In schools the use of the Internet is now common to many classes and homework assignments. However while the relevance of web search results to given keywords has improved substantially over the past decade the appropriateness of the results is uncatered for. On a keyword search for snakes the same results will be given whether the user is a seven year old elementary school kid or a snake expert. Prior work on assessing reading level includes Heilman et al. 2007 who experiment with a system that employs grammatical features and vocabulary to predict readability. The system is part of the the REAP tutor designed to help ESL learners improve their vocabulary skills.