NER is an important subtask in information extraction which suffers severely from the noisy nature and short length of tweets. Toward addressing these problems, Here NER system uses tweet segmentation, POS tagging, global and local context together to identify candidate named entities from the noisy Twitter stream and random walk model to identify final named entities. Also, the system generates a tweet summary for each identified named entity. The experimental results for tweet dataset with tweet segmentation, POS tagging, and different K values prove the effectiveness of this NER system.
| Autorius: | Minal Sonmale |
| Leidėjas: | LAP LAMBERT Academic Publishing |
| Išleidimo metai: | 2020 |
| Knygos puslapių skaičius: | 100 |
| ISBN-10: | 6202524235 |
| ISBN-13: | 9786202524230 |
| Formatas: | Knyga minkštu viršeliu |
| Kalba: | Anglų |
| Žanras: | Archiving, preservation and digitization |
Parašykite atsiliepimą apie „NER for Opinion Summarization Over Twitter Dataset: Named Entity Recognition for Opinion Summarization Using Tweet Segmentation Over Twitter Dataset“
