Customer purchase forecasting for online tourism: A data-driven method with multiplex behavior data

Online tourism has received increasing attention from scholars and practitioners due to its growing contribution to the economy. While related issues have been studied, research on forecasting customer purchases and the influence of forecasting variables, online tourism is still in its infancy. Therefore, this paper aims to develop a data-driven method to achieve two objectives: (1) provide an accurate purchase forecasting model for online tourism and (2) analyze the influence of behavior variables as predictors of online tourism purchases. Based on the real-world multiplex behavior data, the proposed method can predict online tourism purchases accurately by machine learning algorithms. As for the practical implications, the influence of behavior variables is ranked according to the predictive marginal value, and how these important variables affect the final purchase is discussed with the help of partial dependence plots. This research contributes to the purchase forecasting literature and has significant practical implications.     

Balancing risk with virtual private networking during a pandemic

When the pandemic struck and teaching went online worldwide, universities had to make pressing decisions that balanced cybersecurity against other factors, including health and safety, usability, and cost. One such challenge Indiana University (IU) faced was how to accommodate the secure telecommunications needs of 130,000 faculty, staff, and students who would now be teaching, learning, doing research, and working from home. Some universities reflexively promoted virtual private network (VPN) use for all activities. Such an approach would have been unsustainable at IU, however, owing both to the licenses and resources needed for the sheer number of users and to the high-throughput applications on which they rely. Perhaps even worse, it would have increased the chances that the VPN would be unavailable during a critical incident or other situation in which secure communications must be guaranteed. Instead, IU launched an awareness campaign demonstrating exactly when VPN use is and isn’t needed. In addition, network staff employed a VPN feature called split tunneling to reduce the load. This article discusses the advantages and disadvantages of this approach and how IU made the decision to balance both sides of the risk equation to ensure the continued advancement of its mission throughout the pandemic.     

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。     

区域森林公园生态效率实证分析——以华东地区为例

采用Malmquist-Luenberger指数法对华东地区森林公园生态效率值进行测定的结果表明:从时序范围看,2011—2016年华东地区森林公园生态效率的均值为1. 006,但从2015年起进入下滑态势,说明华东地区森林公园的生态效率已经达到有效水平,但在扩大森林公园规模的同时忽视了对环境的有效保护以及资源的合理配置;从空间差异角度看,只有浙江与安徽两省的森林公园生态效率值未达到有效水平(分别为0. 851和0. 871),其余各省(市)的森林公园生态效率值均大于1,说明区域内存在一定的差异性;技术进步指数下降是制约MalmquistLuenberger指数提升的主要原因。由此可见,加大技术投入,促成交流合作机制,形成特有的竞争优势等,对提高区域森林公园生态效率是十分重要的。     

贫困山区农户茶叶生产技术效率及影响因素研究

基于湖北省382份茶叶种植户调研数据,采用随机前沿生产函数模型“一步估计”法分析贫困山区农户茶叶生产技术效率和影响因素。研究结果表明:样本户茶叶种植存在较大改进空间;样本户资本和劳动力投入剩余;茶园的海拔、树龄、茶农是否加入合作社、种植面积、户主是否受过种茶培训等因素对农户茶叶生产技术效率有重要影响。因此,提出紧抓乡村振兴战略机遇,推动贫困山区茶产业高质量发展;建立茶产业“三项补贴”,改善茶园基础设施条件;优化茶叶种植布局,提高合作社服务水平;成立茶产业发展基金,加快培育职业茶农,以提高贫困山区农户茶叶生产技术效率。     

Comparative study of deep learning models for analyzing online restaurant reviews in the era of the COVID-19 pandemic

Online reviews remain important during the COVID-19 pandemic as they help customers make safe dining decisions. To help restaurants better understand customers’ needs and sustain their business under current circumstance, this study extracts restaurant features that are cared for by customers in current circumstance. This study also introduces deep learning methods to examine customers’ opinions about restaurant features and to detect reviews with mismatched ratings. By analyzing 112,412 restaurant reviews posted during January-June 2020 on Yelp.com, four frequently mentioned restaurant features (e.g., service, food, place, and experience) along with their associated sentiment scores were identified. Findings also show that deep learning algorithms (i.e., Bidirectional LSTM and Simple Embedding + Average Pooling) outperform traditional machine learning algorithms in sentiment classification and review rating prediction. This study strengthens the extant literature by empirically analyzing restaurant reviews posted during the COVID-19 pandemic and discovering suitable deep learning algorithms for different text mining tasks.     

基于超效率SBM模型的珠三角区域物流效率及影响因素分析

物流效率是反映物流绩效水平的重要指标.以珠三角城市群9个地市(不含香港特别行政区和澳门特别行政区)为研究对象,基于珠三角9个地市2010—2019年的面板数据,采用超效率SBM模型和GM L指数,测算珠三角城市群各地市的物流效率值及变动趋势.研究表明:技术进步是珠三角区域物流效率提升的主要因素,综合管理水平有待提高;珠三角区域的物流效率总体处于中等偏上效率水平且变化不大,深圳市物流效率水平提升较快且效率最高,东莞市物流效率最低,珠海市效率值起伏最为明显,整个测算期间呈"W"形走势.为进一步探讨外部因素对珠三角区域物流效率的影响程度,选取Tobit模型进行回归分析,发现区位优势对珠三角地区物流效率的提升起到了重要的作用.     

External pressure or internal culture? An innovation diffusion theory account of small retail businesses’ social media use

COVID-19 put unprecedented external pressure on small businesses to adopt or increase use of social media while not all small businesses are internally ready for this rapid change. This study investigated the roles of external pressures and organizational culture of openness and learning in driving small retail business owners'/managers' social media use decisions by impacting their perceived usefulness and barriers, based on the innovation-decision process model from the diffusion of innovations theory and the theory of reasoned action. An online survey with structured measurements was administered to 411 U.S. small retail business owners/managers. Results from structural equation modeling revealed that external pressures positively influenced small business owners'/managers’ perceived usefulness and barriers and social media use intention. In addition, the culture of openness and learning positively influenced the perceived usefulness while mitigating the perceived barriers, thereby directly and indirectly influencing the social media use intention. Theoretical and managerial implications are discussed.     

Deep learning for modeling the collection rate for third-party buyers

This study evaluates a wide range of machine learning techniques such as deep learning, boosting, and support vector regression to predict the collection rate of more than 65,000 defaulted consumer credits from the telecommunications sector that were bought by a German third-party company. Weighted performance measures were defined based on the value of exposure at default for comparing collection rate models. The approach proposed in this paper is useful for a third-party company in managing the risk of a portfolio of defaulted credit that it purchases. The main finding is that one of the machine learning models we investigate, the deep learning model, performs significantly better out-of-sample than all other methods that can be used by an acquirer of defaulted credits based on weighted-performance measures. By using unweighted performance measures, deep learning and boosting perform similarly. Moreover, we find that using a training set with a larger proportion of the dataset does not improve prediction accuracy significantly when deep learning is used. The general conclusion is that deep learning is a potentially performance-enhancing tool for credit risk management.