As students advance to higher grade levels, they learn new words. The documents intended for upper grade levels will contain more advanced vocabularies, reflecting the assumed aptitude level of the intended audience. In this study, we first classified all the words in a pre-labeled document collection into various grade level categories. We then calculated the distribution of words from each grade level for all the documents. The eventual goal of our study is to build a system that automatically assigns the appropriate grade level label to each document in the NSDL repository. This will allow the educators to search more easily for material appropriate to specific audiences.
The available dataset for this study comes from the Eisenhower National Clearinghouse. This dataset contains a total of 8,417 documents with labels specifying the intended grade levels.
Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/105192 |
Date | 04 1900 |
Creators | Fountain, Tony, Moore, Reagan |
Source Sets | University of Arizona |
Language | English |
Detected Language | English |
Type | Report |
Page generated in 0.0014 seconds