Deep learning-based picture sentiment polarity analysis method

Abstract

The invention discloses a deep learning-based picture sentiment polarity analysis method, and relates to the technical field of image content understanding and big data analysis. In a conventional picture sentiment analysis method, the final prediction precision is non-ideal due to simple models and features. At present, a deep learning method is used to perform training in a large-scale training set, but the noises of the training set are excessively high, so that the final performance is limited. In the deep learning-based picture sentiment polarity analysis method, a mode of obtaining data directly from a network is adopted, and slave data scale is large. Only sentiment polarity information of common words needed to be obtained during data preparation is possibly needed to be manually annotated. Later, the whole image obtaining and cleaning work can be automatically finished, so that the required labor cost is very low. In the data obtaining stage, two data cleaning processes are introduced, so that a large portion of noises due to inconsistence of pictures and tags can be eliminated. According to the method, priori knowledge is used for filtering the training set, so that the noises of the training set are reduced; and an improved network structure is used assistantly, so that the picture sentiment prediction accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of image content understanding and big data analysis, in particular to a method for image emotion analysis. Background technique [0002] With the development of the Internet and the popularity of smart phones, social networks have an irreplaceable role in people's daily life. More and more people began to express their opinions through social networking platforms, and a large amount of user-generated data was generated as a result. [0003] User Generated Content (UGC) refers to original content uploaded by users, which originates from users and ultimately serves users. In the era of web2.0, users no longer passively accept Internet content, but participate in it as a subject, and become producers and disseminators in addition to the role of users. [0004] In the face of huge user-generated data, how to effectively use it has become an urgent problem to be solved. Aiming at these data, related research ...

AI Technical Summary

Problems solved by technology

For example: Quanzeng You published the article "Robust Image Sentiment Analysis Using Progressively Trained and Domain Transferred Deep Networks" in 2015, using the Sentibank dataset and using the idea of ​​self-learning to improve the deep learning network and construct a PCNN network, which can resist to a certain extent Noise problem in network data sets, but due to the inherent limitations of self-learning theory, its performance improvement is limited

[0011] To sum up, the traditional image sentiment analysis method requires a small data set, but the final prediction accuracy is not ideal due to the simplicity of the model and features

Some current methods using deep learning are trained in large-scale training sets, but due to the excessive noise of the training set, the final performance is limited

Images