Emotion Analysis Of Wheelchair Online Review Based On Machine Learning

China's population structure is undergoing s ignificant changes,and the aging population is becoming increasingly serious.Many elderly people need wheelchairs to maintain their daily life due to physical reasons or cerebral palsy.As an important medical and health care device,the elderly's demand for the comfort,safety performance and function of wheelchairs is growing day by day,which has also become the focus of the current wheelchair market.How to improve the experience of the elderly has become a difficulty for wheelchair manufacturers.While the market space of wheelchair manufacturers is expanding,how to become the industry leader is also very important.With the advent of the era of big data,the data on the Internet has increased substantially.The main online purchase platform can obtain tens of thousands of users' feedback on a product,with a growing number and more complex content.As a product with more prominent demand in the elderly group,wheelchair has a considerable record of online sales.This paper uses Spyder software on the shopping platform of Jingdong,according to the c lassification standard of manual wheelchair and intelligent wheelchair,crawls the product review information,as the analys is object,to investigate the evaluation of the advantages and disadvantages of the elderly wheelchair.Based on convolution neural network model,word cloud map and LDA topic model,this paper mainly mines the potential information of comments.Firstly,we preprocess the crawled comment data,including word segmentation,adding a custom dictionary and removing the stop words.Then,we construct a CNN convolutional neural network classification model,which divides the comment data into four data sets: positive comment of manual wheelchair,negative comment of manual wheelchair,positive comment of intelligent wheelchair and negative comment of intelligent wheelchair.Then we create a new data set based on these four data sets Word cloud model is used to highlight key words,and LDA theme model is used to evaluate the good and bad of two kinds of wheelchairs,analyze the theme,and further mine the potential information.The results show that the advantages of the manual wheelchair are high safety factor,portability and portability,while the disadvantages are mainly incomplete functions and laborious use;the advantages of the intelligent wheelchair are simple operation,multiple functions,but the disadvantages are low safety factor and inconvenient handling.Finally,it puts forward suggestions and suggestions for the improvement of the wheelchair manufacturers,the intelligent elderly care platform and relevant government departments according to the quality,function,comfort level and safety system of wheelchairs.