Research On The Influence Of Text Features Of Medical Crowdsourcing Questions On The Quantity And Quality Of Doctors’ Responses

The rise of the Internet has made medical crowdsourcing a new method of paramedical assistance.Patients can throw their own problems to the crowdsourcing platform to seek online responses from doctors on the platform.This greatly improves the ability of patients to obtain information and broadens the channels for problem solving.Although there are many researches on crowdsourcing and medical crowdsourcing,there is a lack of research on the impact of the textual factors of medical crowdsourcing on the quantity and quality of doctors’ replies.Therefore,based on the theoretical basis of crowdsourcing participation motivation and participation performance,this study explores the impact of medical crowdsourcing text characteristics of patients asking questions on the number and quality of doctors’ responses in the medical crowdsourcing community,and gives patients the skills to ask questions to obtain more doctors.Attention and higher quality responses,and provide optimization suggestions for the construction of medical crowdsourcing community websites.This research is mainly carried out from the following three aspects:(1)Based on the research of existing crowdsourcing participation motivation and participation influencing factors,combined with the characteristics of the medical crowdsourcing field,the text features of the online medical crowdsourcing platform question are proposed to the number and response of doctors.The impact quality model of reply quality;(2)Collect the data of "one disease,many questions" in the medical crowdsourcing module in the online medical platform "Wei Yi",and perform data processing to obtain empirical analysis variables;(3)on the basis of the above two steps Empirical analysis,using multiple linear regression to verify the impact of the medical crowdsourcing platform problem description text on the number of doctors’ responses and the quality of responses,while taking into account the count attributes of the variables,using Poisson regression and negative binomial regression for robust testing,and finally drawing conclusions.The empirical results show that(1)the positive emotions in the crowdsourcing problem have no significant impact on the number of doctors’ responses,and have a significant positive effect on the quality of doctors’ responses;(2)the comprehensiveness of the description of the medical crowdsourcing problem has a positive effect on the quality of doctors’ responses Impact,but has no effect on the number of doctors’ responses;(3)The length of the text of the description of the medical crowdsourcing problem negatively affects the number of doctors’ responses;the number of specific questions in the medical crowdsourcing problem positively affects the quality of doctors’ responses;(4)Material rewards have a positive effect on the number of doctor responses.Theoretically,this research improves the field of medical crowdsourcing,and the influencing factors of the quantity and quality of doctor’s response.In practice,the results of this research can provide optimization suggestions for the medical crowdsourcing platform to better assist patients to solve health problems.At the same time,the deep neural network text sentiment analysis method adopted in this research provides technical solutions for the subsequent research on text analysis.