DocumentCode
3739362
Title
TextServer: Cloud-Based Multilingual Natural Language Processing
Author
Llu?s Padr?;Jordi Turmo
fYear
2015
Firstpage
1636
Lastpage
1639
Abstract
TextServer is an efficient language analysis platform which offers a variety of robust NLP services for a wide range of languages. Services can be easily accessed via a web interface, where document collections can either be uploaded and sent to batch processing, or processed in real-time via interactive calls to a web-service from any application running on a remote computer or mobile device. The services run on a high-performance-computing cluster (HPCC), and a scheduler takes care of having enough instances loaded to satisfy incoming client requests with no delay caused by initializations. TextServer offers a variety of analysis levels, ranging from tokenization or lemmatization to coreference resolution and semantic graph extraction, and aims to become a useful resource for data mining applications working on monolingual or multilingual textual data which require to obtain structured data from unrestricted text.
Keywords
"Semantics","Natural language processing","Data mining","Web services","Hidden Markov models","Tagging","Registers"
Publisher
ieee
Conference_Titel
Data Mining Workshop (ICDMW), 2015 IEEE International Conference on
Electronic_ISBN
2375-9259
Type
conf
DOI
10.1109/ICDMW.2015.102
Filename
7395876
Link To Document