Hybrid feature-based sentiment strength detection for big data applications

Yanghui RAO, Haoran XIE, Fu Lee WANG, Kin Man POON, Endong ZHU

Research output: Chapter in Book/Report/Conference proceedingChapters

Abstract

In this chapter, we focus on the detection of sentiment strength values for a given document. A convolution-based model is proposed to encode semantic and syntactic information as feature vectors, which has the following two characteristics: (1) it incorporates shape and morphological knowledge when generating semantic representations of documents; (2) it divides words according to their part-of-speech (POS) tags and learns POS-level representations for a document by convolving grouped word vectors. Experiments using six human-coded datasets indicate that our model can achieve comparable accuracy with that of previous classification systems and outperform baseline methods over correlation metrics. Copyright © 2019 Springer Nature Switzerland AG.
Original languageEnglish
Title of host publicationMultimodal analytics for next-generation big data technologies and applications
EditorsKah Phooi SENG, Li-minn ANG, Alan Wee-Chung LIEW, Junbin GAO
Place of PublicationCham
PublisherSpringer
Pages73-91
ISBN (Electronic)9783319975986
ISBN (Print)9783319975979
DOIs
Publication statusPublished - 2019

Citation

Rao, Y., Xie, H., Wang, F. L., Poon, L. K. M., & Zhu, E. (2019). Hybrid feature-based sentiment strength detection for big data applications. In K. P. Seng, L.-M. Ang, A. W.-C. Liew, & J. Gao (Eds.), Multimodal analytics for next-generation big data technologies and applications (pp. 73-91). Cham: Springer.

Fingerprint Dive into the research topics of 'Hybrid feature-based sentiment strength detection for big data applications'. Together they form a unique fingerprint.