Title :
Towards Named Entity Recognition Method for Microtexts in Online Social Networks: A Case Study of Twitter
Author_Institution :
Dept. of Comput. Eng., Yeunnam Univ., Gyeongsan, South Korea
Abstract :
Given a certain question, named entity recognition (NER) methods can be an efficient strategy to extract relevant answers. The goal of this work is to extend NER methods for analyzing a set of micro texts, which are short text on online social media. To do so, we propose two contextual closure properties to discover contextual clusters of micro texts, which can be expected to improve the performance of NER tasks. Experimental results demonstrate the feasibility of the proposed method for extracting relevant information in online social network applications.
Keywords :
data mining; question answering (information retrieval); social networking (online); text analysis; Twitter; microtexts; named entity recognition method; online social media; online social networks; Data mining; Facebook; Feature extraction; Media; Organizations; Twitter; Multiplex social network; Named entity recognition; Network analysis; Social networks;
Conference_Titel :
Advances in Social Networks Analysis and Mining (ASONAM), 2011 International Conference on
Conference_Location :
Kaohsiung
Print_ISBN :
978-1-61284-758-0
Electronic_ISBN :
978-0-7695-4375-8
DOI :
10.1109/ASONAM.2011.39
