Research On SaaS Service User Satisfaction Based On Sentiment Analysis

With the development of information technology and information needs of enterprise,SaaS(Software as a service)become the first choice for business users replaced the traditional software gradually.However,under the new development opportunities,SaaS market is also facing the uneven quality of service,users are unsatisfactory and so on.The existence of the problem seriously affects the healthy development of SaaS market.The research on SaaS user satisfaction is helpful to provide SaaS provider optimization service to promote the healthy and sustainable development of SaaS.The existing SaaS user satisfaction research is more through the sentiment analysis method to do the sentimental polarity of the study,from the overall evaluation of user satisfaction,no SaaS users comment on the emotional layer of emotional analysis.Based on this,this paper uses the sentiment analysis method to carry on the aspect level sentiment analysis to the SaaS user comment,in order to carry on the computation and analysis to the user satisfaction degree of each attribute dimension of SaaS.Firstly,we have constructed the user satisfaction evaluation dimension of SaaS,secondly we have constructed the user satisfaction attribute dictionary of SaaS,and verified the user satisfaction evaluation dimension of SaaS by attribute dictionary,thirdly,we crawling SaaS user comments from https://www.getapp.com/web site,and extraction attribute feature of the SaaS user comments by dependency parsing analysis,and match the extraction results with the SaaS user satisfaction attribute dictionary,In order to search sentiment word and output attribute words sentimental weighted score.Finally,the weighted scores of attribute words and attribute weights are used to get the overall level of SaaS and the user satisfaction of each industry.Based on the analysis results,SaaS users can choose SaaS products to provide some reference,as well as SaaS enterprise optimization services to provide some reference.The full text is divided into six parts,the specific content is as follows:The first part is the introduction.It mainly introduces the research background and the research significance,the research content,the research method and the innovation work done in this paper.The second part is related theory and research status quo.Firstly,we briefly introduction the concept of SaaS and SaaS industry division.Secondly,briefly introduction the concept of SaaS user satisfaction.Thirdly,briefly introduction the concept and application of sentiment analysis method used in this paper.Finally,analysis the SaaS user satisfaction research status,put forward the previous research deficiencies,to determine the SaaS user satisfaction research method.The third part is constructing SaaS user satisfaction evaluation dimension and attribute dictionary.Firstly,based on the analysis of the service quality model and the customer satisfaction index model,according to the existing SaaS customer satisfaction factor and SaaS model,we constructed the SaaS customer satisfaction evaluation dimension.Secondly,based on the SaaS user satisfaction evaluation dimension,we constructed the SaaS user satisfaction attribute dictionary by using the WordNet synonym extension and similarity calculation.to construct the SaaS user satisfaction attribute dictionary.Finally,we use the SaaS user satisfaction attribute dictionary to carry on the hypothesis verification to the proposed user satisfaction evaluation dimensionThe fourth part is based on the online comment of SaaS user satisfaction analysis.Firstly,we collected and preprocessed the SaaS comment data set.Secondly,briefly introduction the dependency analysis and the Sentiment By Term algorithm of the attribute level sentiment analysis method.Finally,calculation the user satisfaction of SaaS and the industry sentimental score.The fifth part is the conclusion and discussion.First of all,comprehensive analysis and summary the attribute level of user satisfaction of SaaS user comments,and analysis the overall user satisfaction with SaaS.Secondly,the detailed analysis of SaaS industry,and the user satisfaction is analyzed from the perspective of industry dimension and attribute dimension respectively.The difference of user satisfaction and the satisfaction degree at the attribute level are different.Finally,according to the analysis results proposed relevant summary recommendations.The sixth part is summary and outlook.This paper summarizes the research,and puts forward the shortcomings of the research and the follow-up research ideas.