Research on image sentiment analysis technology based on sparse representation

Abstract

Many methods based on deep learning have achieved amazing results in image sentiment analysis. However, these existing methods usually pursue high accuracy, ignoring the effect on model training efficiency. Considering that when faced with large‐scale sentiment analysis tasks, the high accuracy rate often requires long experimental time. In view of the weakness, a method that can greatly improve experimental efficiency with only small fluctuations in model accuracy is proposed, and singular value decomposition (SVD) is used to find the sparse feature of the image, which are sparse vectors with strong discriminativeness and effectively reduce redundant information; The authors propose the Fast Dictionary Learning algorithm (FDL), which can combine neural network with sparse representation. This method is based on K‐Singular Value Decomposition, and through iteration, it can effectively reduce the calculation time and greatly improve the training efficiency in the case of small fluctuation of accuracy. Moreover, the effectiveness of the proposed method is evaluated on the FER2013 dataset. By adding singular value decomposition, the accuracy of the test suite increased by 0.53%, and the total experiment time was shortened by 8.2%; Fast Dictionary Learning shortened the total experiment time by 36.3%.

Manifold-based sparse representation for opinion mining

“…The sparse representation by singular vector decomposition is used to enhance the efficiency of deep learning in image sentiment analysis. It ignores the local smoothness assumption as an effective prior knowledge 31 .…”

“…Sparse representation has led to promising results in a wide range of applications, including visual recognition 27 , image synthesis 28 , animation 29 , denoising 30 , etc. Recently, sparse representation is applied to sentiment analysis 31 , 32 . -norm sparsity is exploited to represent micro-videos with the aim of finding the main frame including sentiment 33 .…”

“… -norm sparsity is exploited to represent micro-videos with the aim of finding the main frame including sentiment 33 . The singular vector decomposition is also used for the sparse representation of image sentiments 31 . The success of sparse learning in sentiment analysis on image and video and the importance of graph representation in manifold-based feature extraction are the author’s motivation for incorporating them together.…”

RGB‐guided hyperspectral image super‐resolution with deep progressive learning

“…Traditional methods mathematically model the relationship between LR HS image, HR RGB image and HR HS image, concentrate the prior exploitation of the desired HR HS image, for example, sparse representation [36, 37]. Matrix factorisation (MF) is the most popular technique employed by traditional methods [13–17, 38, 39].…”

Intelligent media computing technology and application for media convergence

“…The third article researches different types of feature selection methods, considering various sparse learning algorithms, followed by a novel fast classification method for fine grained emotion recognition tasks [3]. The authors confirmed that, using this faster dictionary learning algorithm, the convergence of the network model is accelerated, and a better balance between classification efficiency and classification quality is achieved.…”