Detecting Non-covered Questions in Frequently Asked Questions Collections (CROSBI ID 660331)
Prilog sa skupa u časopisu | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Karan, Mladen ; Šnajder, Jan
engleski
Detecting Non-covered Questions in Frequently Asked Questions Collections
Frequently asked questions (FAQ) collections are a popular and effective way of representing information, and FAQ retrieval systems provide a natural-language interface to such collections. An important aspect of efficient and trustworthy FAQ retrieval is to maintain a low fall-out rate by detecting non-covered questions. In this paper we address the task of detecting non-covered questions. We experiment with threshold-based methods as well as unsupervised one-class and supervised binary classifiers, considering tf-idf and word embeddings text representations. Experiments, carried out on a domain-specific FAQ collection, indicate that a cluster-based model with query paraphrases outperforms threshold-based, one-class, and binary classifiers.
FAQ retrieval, Novelty detection, Question answering
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
387-390.
2017.
nije evidentirano
objavljeno
10.1007/978-3-319-59569-6_47
Podaci o matičnoj publikaciji
Lecture notes in computer science
Frasincar, Flavius ; Ittoo, Ashwin ; Nguyen, Le Minh ; Métais, Elisabeth
Liege: Springer
978-3-319-59569-6
0302-9743
Podaci o skupu
International Conference on Applications of Natural Language to Information Systems.
poster
21.06.2017-23.06.2017
Liège, Belgija