One model to rule them all: unified classification model for geotagging websites
Электронный научный архив УРФУ
Информация об архиве | Просмотр оригинала| Поле | Значение | |
| Заглавие |
One model to rule them all: unified classification model for geotagging websites
|
|
| Автор |
Volkov, Alexey
Serdyukov, Pavel Maslov, Michael |
|
| Тематика |
ИНФОРМАТИКА
ИНФОРМАЦИОННЫЙ ПОИСК В ИНТЕРНЕТЕ ПОИСК ИНФОРМАЦИИ В ИНТЕРНЕТЕ КОНФЕРЕНЦИИ GEOTAGGING CLASSIFICATION MODELS ENTITIES MACHINE LEARNING |
|
| Описание |
The paper presents a novel approach to finding regional scopes (geotagging) of websites. It relies on a single binary classification model per region type to perform the multi-label classification and uses a variety of different features that have not been yet used together for machine-learning based regional classification of websites. The evaluation demonstrates the advantage of our one model per region type method versus the traditional one model per region approach.
|
|
| Дата |
2012-11-13T11:18:42Z
2012-11-13T11:18:42Z 2012 |
|
| Тип |
Article
Journal article (info:eu-repo/semantics/article) Published version (info:eu-repo/semantics/publishedVersion) |
|
| Идентификатор |
Volkov A. One model to rule them all: unified classification model for geotagging websites / A. Volkov, P. Serdyukov, M. Maslov // VI Russian Summer School in Information Retrieval, August 6–10, 2012. Proceedings of the Sixth Russian Young Scientists Conference in Information Retrieval / B. Sokolov, P. Braslavski (Eds.). — Yaroslavl, 2012. — P. 45-57.
978-5-8397-0888-4 http://elar.urfu.ru/handle/10995/4560 |
|
| Язык |
en
|
|
| Связанные ресурсы |
RuSSIR 2012
|
|
| Формат |
246241 bytes
application/pdf |
|