Furihata Kenji
Department of Electrical and Electronic Engineering, Faculty of Engineering, Shinshu University, Nagano, Japan.
J Acoust Soc Am. 2008 Dec;124(6):3544-60. doi: 10.1121/1.3003080.
Annoyance to noise is probably a good measure of the potential for complaint and community action. Each individual's degree of annoyance to noise cannot be accurately predicted from conventional laboratory studies and social surveys. Therefore, a logistic prediction model for individual allowable noise levels has been developed. In a previous paper [K. Furihata et al., Acta Acust. Acust. 93, 73-83 (2007)], an annoyance scale using the most frequently used annoyance ratings related to noise for community noise assessment in Japan was constructed. The dose-response relationships were obtained by actually listening to sounds in a field situation (1474 subjects) carried out at 14 metropolitan cities in Japan. The logistic prediction model was developed using the logistic regression parameters estimated from each cumulative percentage curve of seven annoyance scale scores based on all data (n=2940). To verify the hypothesis, individual relationships between annoyance and L(Aeq, (5 min)) for 24 normal-hearing subjects were constructed using a laboratory experiment. The main conclusion drawn from the results is that if at least one dose-response relation is obtained from each subject, the model can predict whether a sound is annoying or not from its L(Aeq,(5 min)) value. Moreover, validation of the model is needed in a field situation.
对噪音的烦恼程度可能是衡量投诉可能性和社区行动潜力的一个良好指标。从传统的实验室研究和社会调查中无法准确预测每个人对噪音的烦恼程度。因此，已经开发了一个用于个人允许噪音水平的逻辑预测模型。在之前的一篇论文[K. Furihata等人，《声学学报》93，73 - 83(2007)]中，构建了一个使用与日本社区噪音评估中最常用的与噪音相关的烦恼评级的烦恼量表。剂量 - 反应关系是通过在日本14个大城市进行的实地情况（1474名受试者）中实际聆听声音获得的。逻辑预测模型是使用基于所有数据（n = 2940）的七个烦恼量表分数的每个累积百分比曲线估计的逻辑回归参数开发的。为了验证该假设，使用实验室实验构建了24名听力正常受试者的烦恼与L(Aeq, (5 min))之间的个体关系。从结果得出的主要结论是，如果从每个受试者至少获得一个剂量 - 反应关系，该模型可以根据其L(Aeq,(5 min))值预测声音是否令人烦恼。此外，需要在实地情况中对该模型进行验证。
