Document Ranking Applied to Second Language Learning

This paper addresses the needs of language learners and teachers by combining keyword-based search and language level information on an algorithm that can rank documents by pertinence to the required topic (keywords) and adequacy to the user’s language level. We conducted several experiments using t...

Full description

Saved in:
Bibliographic Details
Published inAdvances in Information Retrieval pp. 618 - 624
Main Authors Wilkens, Rodrigo, Zilio, Leonardo, Fairon, Cédrick
Format Book Chapter
LanguageEnglish
Published Cham Springer International Publishing
SeriesLecture Notes in Computer Science
Subjects
Document Ranking
Language Learning Framework
Level Language
Linguistic Information
Readability Measures
Online AccessGet full text

Cover

More Information
Summary:This paper addresses the needs of language learners and teachers by combining keyword-based search and language level information on an algorithm that can rank documents by pertinence to the required topic (keywords) and adequacy to the user’s language level. We conducted several experiments using the EF-CAMDAT corpus (annotated for topic and level) and we observed that the best ranking results were an average of BM25 and linguistic information. We also saw that the grammar of level C1 is the best indicator for level. Finally, we proposed a customization for prioritizing beginner or intermediate levels at the top of the rank.
ISBN:9783319769400
3319769405
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-319-76941-7_53