Visual and Textual Sentiment Analysis of a Microblog Using Deep Convolutional Neural Networks

Yuhai Yu,Jiana Meng,Zhehuan Zhao,Hongfei Lin

doi:10.3390/a9020041

Yuhai Yu, Jiana Meng + Show 2 more

Open Access

https://doi.org/10.3390/a9020041

Copy DOI

Abstract

Sentiment analysis of online social media has attracted significant interest recently. Many studies have been performed, but most existing methods focus on either only textual content or only visual content. In this paper, we utilize deep learning models in a convolutional neural network (CNN) to analyze the sentiment in Chinese microblogs from both textual and visual content. We first train a CNN on top of pre-trained word vectors for textual sentiment analysis and employ a deep convolutional neural network (DNN) with generalized dropout for visual sentiment analysis. We then evaluate our sentiment prediction framework on a dataset collected from a famous Chinese social media network (Sina Weibo) that includes text and related images and demonstrate state-of-the-art results on this Chinese sentiment analysis benchmark.

Highlights

As the number of webcams has increased, more and more people enjoy posting their experiences in opinionated texts and images using audio or video and expressing their opinions about all sorts of events and subjects in online social networks
We have described two experiments with convolutional neural networks built on top of pre-trained vectors trained by word2vec
We leverage recently developed deep learning models to extract both textual and visual features to analyze the sentiment expressed in Chinese microblogs

Summary

Introduction

As the number of webcams has increased, more and more people enjoy posting their experiences in opinionated texts and images using audio or video and expressing their opinions about all sorts of events and subjects in online social networks. Social networks such as Twitter and Sina. We focus on automatically detecting the sentiments expressed in these large-scale datasets. Extracting sentiment is both meaningful and a major challenge for many social media analytics tasks such as advertising or recommendations. We can understand the sentiment in a message from its brief textual component and from the visual content provided by the user

Methods

Results

Conclusion