基于大规模街景图片人工智能分析的精细化城市修补方法研究

在中国快速城镇化阶段“重量”而“轻质”的建设过程中，城市建成区尤其是老城区的景观环境产生了诸如景观视廊受阻、风貌破败等大量问题。对此，自2015年中央城市工作会议以来，各地相继开展了城市修补专项规划，以修复及更新城市建成环境，促进城市空间品质提升。然而，由于缺乏对城市环境整体效应的统筹考虑，故在城市实际修补过程中，城市局部地块的品质提升反而对城市整体品质构成负面影响。因此，在人工智能及大数据技术深度介入城市规划和设计实践探索的基础上，研究基于全卷积神经网络模型(FCN)和城市场景要素深度学习数据集，对城市景观环境中的各要素进行了大规模且高颗粒度的精确识别，同时与空间数据叠加，对复杂建成环境中的景观问题进行精确分析，并基于分析成果辅助后续城市规划设计实践，逐层递进地对城市复杂建成环境进行精细化修补。选择位于嵩山脚下的登封市作为案例，探索人工智能技术在辅助城市修补等规划领域的前瞻性应用。     

The effect of environmental regulations,top management commitment,and organizational learning on green product innovation: Evidence from automobile industry

This paper proposes and empirically examines a model to investigate the effect of environmental regulations, top management commitment (TMCO) and organizational learning toward green product innovation (GPI). The proposed theoretical model, grounded in dynamic capabilities view (DCV) and upper echelons theory, is analyzed by Partial least squares (PLS) method using the data from Indian automotive manufacturing firms. The findings indicate the importance of TMCO and organizational learning for implementing GPI (in response to regulations), and achieve desired performance. Further, organizational learning fully mediates between commitment of top management and GPI. The findings can be useful for managers in automotive manufacturing firms who are interested toward implementing GPI. The paper contributes to green innovation literature by empirically examining the role of TMCO and organizational learning for GPI.     

A machine-learning method for improving crash injury severity analysis: a case study of work zone crashes in Cairo,Egypt

Abstract

The quality of vehicular collision data is crucial for studying the relationship between injury severity and collision factors. Misclassified injury severity data in the crash dataset, however, may cause inaccurate parameter estimates and consequently lead to biased conclusions and poorly designed countermeasures. This is particularly true for imbalanced data where the number of samples in one class far outnumber the other. To improve the classification performance of the injury severity, the paper presents a robust noise filtering technique to deal with the mislabels in the imbalanced crash dataset using the advanced machine learning algorithms. We examine the state-of-the-art filtering algorithms, including Iterative Noise Filtering based on the Fusion of Classifiers (INFFC), Iterative Partitioning Filter (IPF), and Saturation Filter (SatF). In the case study of Cairo (Egypt), the empirical results show that: (1) the mislabels in crash data significantly influence the injury severity predictions, and (2) the proposed M-IPF filter outperforms its counterparts in terms of the effectiveness and efficiency in eliminating the mislabels in crash data. The test results demonstrate the efficacy of the M-IPF in handling the data noise and mitigating the impacts thereof.     

Object detection based on visual memory: a feature learning and feature imagination process

ABSTRACT

Visual memory plays an important role for the human’s visual system to detect objects. The features of an object stored in the visual memory have much lower dimensions than the features contained within an image. We simulate the visual memory as a feature learning and feature imagination (FLFI) process to build an object detection algorithm. The method is constructed by a bottom-up feature learning and a top-down feature imagination. The proposed object detection method is tested using publicly available benchmark data sets, and the result indicates that it is fast and more robust.     

深度学习技术在海关风险甄别中的应用研究

随着互联网的高速发展和人工智能时代的到来,越来越多从前必须由人脑完成的工作能够利用计算机技术来完成,而深度学习的出现更解决了传统机器学习算法在计算机视觉领域、自然语言处理领域表现不佳的问题,使机器也能够拥有准确感知图像和语音的能力。人脸识别是深度学习网络最常见的应用场景之一,具有自然、直接、方便的特点,且不需要检测对象配合,因此非常适合用于公共安全领域的风险检测。研究充分结合海关实际需求,搭建基于深度学习技术的人脸识别模型,提供对通关旅客进行实时风险甄别的解决方案,以及海关通关风险防控场景的理论参考,为后续深度学习技术在海关业务的研究提供支撑。     

Structural breaks in the mean of dividend-price ratios: Implications of learning on stock return predictability

In their out-of-sample predictions of stock returns in the presence of structural breaks, Lettau and Van Nieuwerburgh (2008) implicitly assume that economic agents’ perception of the regime-specific mean for the dividend-price ratio is time-invariant within a regime. In this paper, we challenge this assumption and employ least squares learning with constant gain (or constant-gain learning) in estimating economic agents’ time-varying perception for the mean of dividend-price ratio. We obtain better out-of-sample predictions of stock returns than in Lettau and Van Nieuwerburgh (2008) for both the U.S. and Japanese stock markets. Our empirical results suggest that economic agents’ learning plays an important role in the dynamics of stock returns.     

Bankruptcy prediction for small- and medium-sized companies using severely imbalanced datasets

Bankruptcy prediction is still important topic receiving notable attention. Information about an imminent bankruptcy threat is a crucial aspect of the decision-making process of managers, financial institutions, and government agencies. In this paper, we utilize a newly acquired dataset comprising financial parameters derived from the annual reports of small- and medium-sized companies. The data, which reveal the true ratio between bankrupt and non-bankrupt companies, are severely imbalanced and only contain a small fraction of bankrupt companies. Our solution to overcome this challenging scenario of imbalanced learning was to adopt three one-class classification methods: a least-squares approach to anomaly detection, an isolation forest, and one-class support vector machines for comparison with conventional support vector machines. We provide a comprehensive analysis of the financial attributes and identify those that are most relevant to bankruptcy prediction. The highest prediction performance in terms of the geometric mean score is 91%. The results are validated on two datasets from the manufacturing and construction industries.     

航班计划优化研究概述

从航班计划优化的不同时间阶段分析,可以将航班计划优化分为航班计划静态编排优化、基于航班延误预测的航班计划动态反馈优化和基于机场协同决策(A-CDM)的航班计划动态调整角度三类;进而从航班时刻、机型指派、航班频率等编制环节分析了航班计划静态编排优化;随后利用延误波及预测与数据挖掘预测的优化方法分析了基于航班延误预测的航班计划动态反馈优化的相关研究。最后,根据航班计划优化复杂性分析,给出了航班计划优化的发展趋势和未来研究方向。     

Exploring the multi-dimensionality of authenticity in dining experiences using online reviews

The quest for authenticity in dining experiences has become increasingly important. This paper explores authenticity dimensions that are of value to customers in dining experiences, and by that gains a multi-dimensional understanding of authenticity in this context. Following an integrated learning approach using text mining and classification techniques, this paper explores and confirms different dimensions of authenticity by identifying and classifying authenticity judgements in online restaurant reviews. The results suggest that authenticity is a multi-dimensional concept encompassing Authenticity of the Other, Authenticity of the Producer, and Authenticity of the Self as first-level dimensions. Additionally, besides historical and categorical authenticity which have been previously explored in the literature, a new type of authenticity - Deviated Authenticity - emerged as a second-level dimension falling under Authenticity of the Other. This paper enhances existing conceptualisations of authenticity and establishes avenues for exploring the multi-dimensionality of other consumer research concepts using user-generated content.