.png){kind=icon}
.png){kind=icon}
Return
A consumer emotion analysis system based on support vector regression model
.png){kind=icon}
OA
AI
DOI：10.7717/peerj-cs.1381.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The effective means to stimulate economic growth is to enhance consumers' consump-tion capacity. Because many consumers have different consumption habits, they will pay different attention to products. Even the same consumer will have different shopping experiences when buying the same product at different times. By mining the online comments of consumers on the online fitness platform, we can find the characteristics of fitness projects that consumers care about. Analyzing consumers' emotional tendencies towards the characteristics of fitness programs will help online fitness platforms adjust the quality and service direction of fitness programs in a timely manner. At the same time, it can also provide purchase advice and suggestions for other consumers. Based on this goal, this study uses an optimized support vector regression (SVR) model to build a consumer sentiment analysis system, so as to predict the consumer's willingness to pay. The optimized SVR model uses the region convolution neural network (RCNN) to extract features from the dataset, and uses feature data to train the SVR model. The experimental results show that the SVR model optimized by RCNN is more accurate. The improvement of the accuracy of consumer sentiment analysis can accurately help businesses promote and publicize, and increase sales. On the other hand, the increase in the accuracy of emotion analysis can also help users quickly locate their favorite fitness projects, saving browsing time. To sum up, the emotional analysis system for consumers in this paper has good practical value.
Keywords:
Emotional analysis
Willingness to pay
SVR
RCNN
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
2.5
Papers:
3.4K
Citations:
6.9K