Sentiment Analysis for Images on Microblogging by Integrating Textual Information with Multiple Kernel Learning
Image is one of the most important means to express users’ emotions on microblogging, like Sina Weibo. More and more people post only images on it, due to the fast and convenient nature of image. Taking a post only using images on microblogging has been a
- PDF / 611,066 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 21 Downloads / 218 Views
State Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing 210023, China [email protected], [email protected], [email protected] 2 School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China [email protected]
Abstract. Image is one of the most important means to express users’ emotions on microblogging, like Sina Weibo. More and more people post only images on it, due to the fast and convenient nature of image. Taking a post only using images on microblogging has been a new tendency. Most existing studies about sentiment analysis on microblogging focus on the text, or integrate image as an auxiliary information into text, so they are not applicable in this scenario. Although a few methods related to sentiment analysis for image have been proposed, most of them either ignore the semantic gap between low-level visual features and higher-level image sentiments, or require a lot of textual information in the phases of both training and inference. This paper proposes a new sentiment analysis method based on Simple Multiple Kernel Learning (SimpleMKL). Specifically, textual information as a sort of sufficiently emotional source data, we can use it to promote the ability via SimpleMKL to classify images. And once we get the image classifier, none of texts are needed when predicting other unlabelled images. Experimental results show that our proposed method can improve the performance significantly on data we crawled and labelled from Sina Weibo. We find that our method not only outperforms some common methods, like SVM, Naive Bayes, KNN, Random Forest, Adaboost, etc., using the image features of colour, hog, texture, but also outperforms some state-of-the-art methods. Keywords: Sentiment analysis Multiple Kernel Learning
1
·
Microblogging
·
Image sentiment
·
Introduction
Microblogging, such as Twitter, Facebook and Sina Weibo, is a popular social media where millions of people express their feelings, emotions, and attitudes every day. Users can post short text messages, images and other type of c Springer International Publishing Switzerland 2016 R. Booth and M.-L. Zhang (Eds.): PRICAI 2016, LNAI 9810, pp. 496–506, 2016. DOI: 10.1007/978-3-319-42911-3 41
Sentiment Analysis for Images with Multiple Kernel Learning
497
information (multi-modality) through various application software on different devices, such as laptops, mobile phones and tablets. With the rapid development of the Internet, microblogging, served as platform to connect people with each other, has been an indispensable part of the modern society. Although mining sentiment information from microblogging had been researched many years and achieved many great breakthroughs, most contributions only focused on texts [1–4]. Most of researchers think that image can be seen as a bag, like text, made up of plentiful visual words. Thus it also contains a lot of sentiment information. More and more researchers try to integrate text an
Data Loading...
