Image and Textual Sentiment Analysis of Memes

Image and Textual Sentiment Analysis of Memes

Author Name : Thilagavathi G, Kamesh M, Boobala Vignesh P

ABSTRACT

Image and textual sentiment analysis is raising more and more attention with the increasing tendency to express emotions through images and text description. Recently social media users are using images and videos more to express their opinions and share their experiences. Sentiment analysis of such large scale visual content can help in better extracting user sentiments toward events or topics from those memes. While this technology is being progressing, little research has been pivoted on the sentiments in memes. In this project an image and text sentiment prediction framework is built by RNN(Recurrent Neural Network),Inception V3 and Fasttext using transfer learning. We have use two approches to gain better accuracy from the given image and text. In textual analysis of sentiment we use NLP (Natural Language Processing) technique. It is useful to know the kind of sentiment expressed in text description in our dataset. Also we use Multimodal data processing technique to bridge the gap between the text-data and image-data. In this work produce a better accuracy for Memes data compare than the previous state of art studies.

Keywords: Sentimental Analysis, memes, optical character recognition, image captioning.