当前位置: X-MOL 学术Inf. Technol. Manag. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Factors dominating individual information disseminating behavior on social networking sites
Information Technology and Management ( IF 2.3 ) Pub Date : 2017-06-29 , DOI: 10.1007/s10799-017-0278-8
Juan Shi , Kin Keung Lai , Ping Hu , Gang Chen

Identifying dominating features that affect individual information retweeting behavior on social networking sites (SNSs) is crucial to understanding individual retweeting behaivor and developing effective marketing strategies on SNS. However, there is little agreement on what factors are dominating individual information disseminating behavior on SNS, and what’s worse, more and more factors are added into the prediction model, without examining the relevance of them and even why these factors are added is rarely discussed. This leads to undesirable outcomes such as increasing the cost of measuring and computing irrelevant/redundant features. Most importantly, it hinders us from understanding what discriminative features are affecting individual information disseminating behavior. Using a unique real-life Twitter data set consisting of 55,575 twitterers and 9,440,321 tweets, the authors examine what discriminative features are dominating individual information disseminating behavior. The results indicate that topic distance is the most discriminative factor, highlighting that self-presentation motives play an important role in information disseminating decisions. Besides, the amount of information, social relationship and the popularity of the tweet also contribute to individual information disseminating decisions. Experiments demonstrate that adopting only dominating factors can improve prediction performance in terms of various indicators, compared with adopting the full features set. Finally, we conclude the paper by discussing theoretical and practical implications of our findings.

中文翻译:

主导个人信息在社交网站上传播行为的因素

识别影响社交网站(SNSs)上个人信息转发行为的主导功能对于理解个人转发行为和制定有效的SNS营销策略至关重要。但是,关于哪些因素主导了SNS上的个人信息传播行为,人们几乎没有共识,更糟糕的是,越来越多的因素被添加到预测模型中,而没有检查它们的相关性,甚至很少讨论为什么增加这些因素。这导致不良结果,例如增加测量和计算不相关/冗余特征的成本。最重要的是,它使我们无法理解哪些歧视性特征正在影响个人信息传播行为。使用由55个组成的独特的真实Twitter数据集,575条推特和9,440,321条推文中,作者研究了哪些歧视性特征主导着个人信息传播行为。结果表明,主题距离是最有区别的因素,突显了自我表达动机在信息传播决策中起着重要作用。此外,信息量,社交关系和推文的受欢迎程度也有助于个人信息传播决策。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。作者研究了哪些歧视性特征主导着个人信息传播行为。结果表明,主题距离是最有区别的因素,突显了自我表达动机在信息传播决策中起着重要作用。此外,信息量,社交关系和推文的受欢迎程度也有助于个人信息传播决策。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。作者研究了哪些歧视性特征主导着个人信息传播行为。结果表明,主题距离是最有区别的因素,突显了自我表达动机在信息传播决策中起着重要作用。此外,信息量,社交关系和推文的受欢迎程度也有助于个人信息传播决策。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。强调自我表达动机在信息传播决策中起着重要作用。此外,信息量,社交关系和推文的受欢迎程度也有助于个人信息传播决策。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。强调自我表达动机在信息传播决策中起着重要作用。此外,信息量,社交关系和推文的受欢迎程度也有助于个人信息传播决策。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。实验表明,与采用完整功能集相比,仅采用主要因素可以提高各种指标的预测性能。最后,我们通过讨论研究结果的理论和实践意义来结束本文。
更新日期:2017-06-29
down
wechat
bug