联系我们

论文查询结果	返回搜索

论文编号：	7400
作者编号：	2120132351
上传时间：	2015/6/9 13:38:30
中文题目：	翻译型众包竞赛中参与者中标影响因素的实证研究
英文题目：	An Empirical Research on Factors Influencing Winning of Participators in Translation Crowdsourcing Contests
指导老师：	侯文华
中文关键字：	众包竞赛，中标率，参与历史，参与策略
英文关键字：	Crowdsourcing contest, Winning probability, Participation history, Participation strategy
中文摘要：	经济全球化的发展使得企业间的竞争越来越激烈，在买方市场下，企业不得不寻觅新的商业模式，提高运作效率，控制成本。也正是在这种情况下，众包（Crowdsourcing）应运而生并快速发展，众包任务和机制设计问题也受到了国内外学者的关注。现有文献大都从任务发布者和众包平台的角度出发，研究参与者动机及激励机制，基于参与者角度的研究较少，并且研究内容比较单一。因此，本文试图从参与者角度出发，综合考虑参与者中标的影响因素，为参与者及众包平台提供参考。 本文在梳理相关文献的基础上，整合了任务属性、参与历史和参与策略三方面因素，构建了翻译型任务中参与者中标影响因素的Logistic模型，重点研究了参与历史和参与策略对参与者中标的影响；另外，本文还构建了参与者中标调节因素的Logistic模型，进一步研究参与近因对上次中标与此次中标之间关系的调节作用以及任务属性对提交顺序比与此次中标之间关系的调节作用。使用到到网与任务中国上合作发布的系列酒店点评翻译任务（共1279条）的相关数据进行实证研究，得到如下结论： （1）任务难度可以降低参与者中标率，任务竞争强度则相反；参与者以往的中标率对未来任务是否中标具有很好的预测作用，但参与者上次中标会降低此次中标的概率；参与者提交顺序比与其中标概率呈倒U型关系，提交太早或太晚都不利于中标。 （2）参与近因可以加强上次中标对此次中标的负向影响，即两次参与之间的时间间隔越长，上次中标对此次中标的负向影响越显著；任务竞争强度和任务难度对提交顺序比与此次中标之间关系均具有显著的调节作用，且随着任务竞争强度和难度的增加，最优的提交顺序比均会增大。 本文研究结论对参与者选取最优策略、提高中标率具有指导意义，也为众包平台增强用户活跃度、激励用户持续参与提供了参考。
英文摘要：	The expansion of economic globalization makes the competition between enterprises increasingly fierce, thus in order to improve operational efficiency and control costs, the enterprises in a buyer’s market have to figure out a new business model, which results in the rise of crowdsourcing. Since crowdsourcing is booming, researches on crowdsourcing tasks and its mechanism design become globally hot. Most existing literature of the motivation and incentive mechanism of participants were written from the view of task publishers and crowdsourcing platform instead of participants. Therefore, this dissertation made comprehensive consideration of influential factors of winning from the perspective of participants so that to benefit participants and crowdsourcing platform to some extent. Based on the review of relevant literature, task attributes, participation historical factors and strategies were integrated as influential factors and a Logistic model of factors influencing winning of participators in Translation was built, focusing on the influence of historical records and strategies of participation on winning; furthermore, a Logistic model of factors adjusting winning of participators was also built to analyze the moderating effect of participation recency on the relationship between two consecutive winnings as well as the adjustment effect of task attributes on the relationship between summation order ratio and winning. Using the relating data of the 1279 tasks which comes from the cooperation of Daodao.com with Taskcn.com, the conclusion is as following. First, task difficulty could reduce the winning probability, whereas competition intensity could improve it; the historical winning probability could be a reasonably good predictor for the winning in the future, specifically, the last winning could decrease the probability of winning in the next bidding; submission order ratio presented an inverted U-shape relationship with winning probability, namely, either too early or too late submission could be disadvantageous to winning. Second, participation recency could enhance the negative effect of the last winning on the winning probability in the following bidding, i.e. the longer time interval between participations led to more significant negative effect; the relationship between submission order ratio and winning could be adjusted by competition intensity and task difficulty, more precisely, the optimal submission order ratio could be improved with the rise of competition intensity and task difficulty. The conclusion above can help crowdsourcing participators choose best strategy and improve winning ratio, while can be a reference for platform to enhance user activity and continuous participation.
查看全文：	预览  下载(下载需要进行登录)