当前位置: X-MOL 学术Pattern Anal. Applic. › 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.)
Analysing the intermeshed patterns of road transportation and macroeconomic indicators through neural and clustering techniques
Pattern Analysis and Applications ( IF 3.7 ) Pub Date : 2020-02-12 , DOI: 10.1007/s10044-020-00872-x
Carlos Alonso de Armiño , Miguel Ángel Manzanedo , Álvaro Herrero

As is widely acknowledged, the transportation of goods by road can, in one way or another, be linked to a range of macroeconomic indicators. A hybrid artificial intelligence system is proposed in this paper to analyse the interaction between transportation patterns and the economy. The temporal patterns of road transportation and macroeconomic trends are studied, by combining the use of both (supervised and unsupervised) neural networks and clustering techniques. The proposed system is validated, by establishing links between road transportation data from Spain and macroeconomic trends over 6 years (2011–2017). The results reveal an interesting inner structure of the data, through data visualizations of intermeshed relations between road transportation patterns and macroeconomic indicators. The same data structure was also visible in the output of the clustering techniques. Furthermore, a number of high-quality predictions were advanced by processing the road transportation data as time series, and forecasting the future values of the main series. These results demonstrated the validity of the proposed linkage between road transportation data and macroeconomic indicators.

中文翻译:

通过神经和聚类技术分析公路运输与宏观经济指标的相互联系的模式

众所周知,公路运输可以以一种或另一种方式与一系列宏观经济指标联系起来。本文提出了一种混合人工智能系统,以分析运输方式与经济之间的相互作用。通过结合使用(有监督和无监督)神经网络和聚类技术,研究了道路运输的时间模式和宏观经济趋势。通过在西班牙的道路运输数据与过去6年(2011-2017年)的宏观经济趋势之间建立联系,对提议的系统进行了验证。通过对道路运输模式与宏观经济指标之间相互联系的关系进行数据可视化,结果揭示了数据的有趣内部结构。在聚类技术的输出中也可以看到相同的数据结构。此外,通过将道路运输数据作为时间序列处理并预测主序列的未来值,提出了许多高质量的预测。这些结果证明了道路运输数据与宏观经济指标之间建议的联系的有效性。
更新日期:2020-02-12
down
wechat
bug