基于AdaBoost的电信客户流失预测模型

随着电信业改革的深入和竞争的不断加剧,各大电信企业的客户流失率逐步攀升。在深入分析电信业客户流失问题的基础上,针对目前研究多采用单分类器模型的不足,提出了一种基于组合分类器的电信客户流失预测模型。实证结果表明该模型能有效提升预测准确率,为今后的研究提供了新的研究思路。     

基于LVQ神经网络的电信企业客户流失预测模型研究

随着人工智能的快速发展,人工神经网络被广泛地运用到分类预测领域。文章首先明确了客户流失的定义及其分类,然后分析了LVQ神经网络的基本原理,最后从研究样本的确定、预测变量的选取、模型的训练及评估三个方面构建了基于LVQ神经网络的电信企业客户流失预测模型,以期为电信企业客户流失预测模型的设计提供一定的借鉴意义。     

Boosting算法在某移动通信公司客户离网预测中的应用研究

在对某移动通信公司客户离网深入分析基础上,针对以往研究采取的算法精度不高的局限性,引入Boosting算法,进行客户离网预测研究。实验结果表明,文中所建模型具有较强的稳定性,且各个指标值都能达到较高水平。     

基于自组织数据挖掘的电子商务客户流失预测模型

为解决电子商务客户流失预测中的高维、非线性问题,本文将自组织数据挖掘理论(SODM)引入客户流失预测,提出一种新颖的基于自组织数据挖掘的电子商务客户流失预测模型。该方法将自组织数据挖掘中的客观系统分析算法(OSA)和改进分组数据处理网络(GMDH)集成起来进行电子商务客户流失预测。首先利用OSA算法选择出重要的电子商务客户流失关键属性,然后将训练样本送入改进GMDH网络进行学习和训练,进而对测试样本客户流失状态进行判别。将该方法应用于某网上商店客户流失预测实证分析,预测结果验证了该方法对包含多种因素影响的电子商务客户流失预测具有优势,基于自组织数据挖掘的电子商务客户流失预测模型具有较强的实用性和可操作性。     

基于BP神经网络的ETC系统客户的流失分析研究

文章应用神经网络数据分析技术研究ETC系统客户流失的状况,详细研究了如何建立ETC系统客户流失基本模型。通过对客户的基本数据进行神经网络预测,可以发现描述流失客户基本特征的属性值集合以及对应的是否流失的结论。文章给出的是改进的神经网络的预测方法,可提高BP神经网络的收敛速度,增强网络的泛化能力,获得了很好的效果。     

客户终生价值测度模型的应用研究——基于大连友好商城的案例分析

本案例研究以客户终生价值测度模型为基本工具,通过挖掘客户历史购买信息,在预测客户未来购买情况的基础上对客户终生价值进行测度,探索其对企业实践的管理意义.以大连友好商城为案例,针对其当前存在的问题,探讨并选择合适的客户终生价值测度模型;通过Pareto/NBD模型预测客户未来的购买次数,Gamma-Gamma模型预测未来的平均购买金额,在此基础上计算客户终生价值;对商城现有客户进行分类,深入分析了不同类别客户的购买特征,提出相应的差异化营销策略.研究表明,通过测度客户终生价值,可以帮助企业识别并分类客户,进而实施差异化的营销策略,提升企业竞争优势.     

从客户流失到客户忠诚：对客户竞争力、客户熵和客户型企业的研究

客户流失研究的疗伤范式难以突破“瓶颈”。本研究以预防范式作为理论起点,先后分析了客户演化、客户竞争力、客户熵和客户型企业,得出通过运用客户竞争力,客户型企业能够获得持续的负熵流,客户作为企业网络的核心部分与企业共同创造价值,在多层次沟通和互动的基础上实现终生购买。客户流失从而转化成客户忠诚。     

基于Stacking模型融合的用户购买行为预测研究

在大数据时代背景下,如何利用大量的销售数据精准预测顾客未来需求,成为企业制定客户管理和库存管理决策的一个重要问题。目前关于用户购买行为预测的研究中很少能够预测用户具体的购买时间。基于已有的销售数据,提出了基于机器学习和Stacking集成的综合预测模型预测用户的购买行为,即未来是否购买及其购买时间。将模型应用在一家大型连锁零售企业的需求预测中,并对方法的有效性进行评估。结果表明,基于Stacking集成的融合模型对预测用户未来是否购买具有最佳性能,准确率达85%,AUC值达到0.928;LightGBM集成算法在预测用户购买时间时具有最优性能,相比于融合模型提升了5.5%的预测性能;融合模型+LightGBM算法的组合相比于均使用融合模型提升了9.4%的预测性能。     

多层前馈神经网络在客户流失分析中的应用

本文研究了多层前馈神经网络原理及其后向传播算法,然后结合一个实例构建了客户流失分析的多层前馈神经网络模型,实验表明将该模型用于客户流失分析是可行的.     

Extending monitoring methods to textual data: a research agenda

Textual data has become increasingly common in business analytic data sets. While concept-based text mining offers a method of extracting meaningful information from text data, methods for monitoring of customer perceptions of business processes and products that are discussed in customer-generated documents are not immediately available. We explore the results of two text-mining algorithms and review issues observed in the data that affect uploading the results onto a newly proposed methodological monitoring platform analogous to statistical process control charts. Finally, we discuss several topics for future research in text mining.     

行业资产的Black-Litterman模型配置研究——基于社交网络情绪文本挖掘算法

研究目标:构建反映行业股价走势的基于社交网络文本挖掘算法的行业投资者情绪指标，并改善嵌入行业投资者情绪指标的Black-Litterman模型对资产的配置结果。研究方法:基于社交网络文本挖掘算法度量投资者情绪，运用主成分分析法构建行业投资者情绪指标，并嵌入Black-Litterman模型中构建投资者观点矩阵，确定行业资产配置比。研究发现:基于行业投资者情绪的BL模型有效提高了资产配置的日均收益率和夏普比率。实证结果在样本外验证(除受新冠疫情影响阶段)、暴涨暴跌阶段以及经过允许卖空和交易成本调整后仍稳健，进而证实了投资者情绪对资产组合有显著影响。研究创新:基于社交网络文本挖掘算法构建投资者情绪指数，解决了仅依赖于预期收益或历史数据的预测模型无法直观揭示投资者心理认知和行为的局限性问题，从一个崭新的视角科学地解决Black-Litterman模型中投资者观点的生成问题。研究价值:扩展了Black-Litterman模型理论体系研究，并推动了行为金融理论在资产配置中的应用。     

Does Service Bundling Reduce Churn?

We examine whether bundling in telecommunications services reduces churn using a series of large, independent cross sections of household decisions. To identify the effect of bundling, we construct a pseudo‐panel dataset and utilize a linear, dynamic panel‐data model, supplemented by nearest‐neighbor matching. We find bundling does reduce churn for all three “triple‐play” services. The effect is only “visible” during times of turbulent demand. We also find evidence that broadband was substituting for pay television in 2009. This analysis highlights that bundling helps with customer retention in service industries, and may play an important role in preserving contracting markets.     

An investigation of social media data during a product recall scandal

As social media has become an important part of modern daily life, users often share product opinions online and these tend to spike when large companies undergo crises. This paper investigates customer online responses to a large company crisis by uncovering hidden insights in social media comments and presents a framework for handling social media data and crisis management. Analysis of textual Facebook data from users responding to the 2013 horsemeat scandal is presented. In this study, we used a novel comprehensive data analysis framework alongside a text-mining framework to objectively classify and understand customer perceptions during this horsemeat scandal. This framework provides an effective approach for investigating customer perception during a company crisis and measures the effectiveness of crisis management practices which the company has adopted. Our analyses show that social media can provide important insights into customer behaviour during crisis communications.     

New perspectives on forecasting inflation in emerging market economies: An empirical assessment

We use a broad-range set of inflation models and pseudo out-of-sample forecasts to assess their predictive ability among 14 emerging market economies (EMEs) at different horizons (1–12 quarters ahead) with quarterly data over the period 1980Q1-2016Q4. We find, in general, that a simple arithmetic average of the current and three previous observations (the RW-AO model) consistently outperforms its standard competitors—based on the root mean squared prediction error (RMSPE) and on the accuracy in predicting the direction of change. These include conventional models based on domestic factors, existing open-economy Phillips curve-based specifications, factor-augmented models, and time-varying parameter models. Often, the RMSPE and directional accuracy gains of the RW-AO model are shown to be statistically significant. Our results are robust to forecast combinations, intercept corrections, alternative transformations of the target variable, different lag structures, and additional tests of (conditional) predictability. We argue that the RW-AO model is successful among EMEs because it is a straightforward method to downweight later data, which is a useful strategy when there are unknown structural breaks and model misspecification.     

Criminal incident prediction using a point-pattern-based density model

Law enforcement agencies need crime forecasts to support their tactical operations; namely, predicted crime locations for next week based on data from the previous week. Current practice simply assumes that spatial clusters of crimes or “hot spots” observed in the previous week will persist to the next week. This paper introduces a multivariate prediction model for hot spots that relates the features in an area to the predicted occurrence of crimes through the preference structure of criminals. We use a point-pattern-based transition density model for space–time event prediction that relies on criminal preference discovery as observed in the features chosen for past crimes. The resultant model outperforms the current practices, as demonstrated statistically by an application to breaking and entering incidents in Richmond, VA.     

A replication and extension of quality management into the supply chain

As competition moves beyond a single firm into the supply chain, researchers are beginning to explore quality management (QM) in a supply chain context. The literature suggests that supply chain management (SCM) consists of internal practices, which are contained within a firm, and external practices, which cross organizational boundaries integrating a firm with its customers and suppliers. Supplier quality management and customer focus are two QM practices that are also clearly in the domain of SCM. In this study we investigate how these two supply chain management-related quality practices lead to improved performance and examine the practices that precede and mediate those relationships. In doing so, we replicate and extend the relationships among the QM practices and their effects on firm performance suggested in Kaynak [Kaynak, H., 2003. The relationship between total quality management practices and their effects on firm performance. Journal of Operations Management 21, 405–435] using survey data gathered from firms operating in the U.S. The inclusion of customer focus and supplier quality management in the QM model supports the importance of internal and external integration for quality performance. Implications of the results for researchers and practitioners are discussed, and further research implications are suggested.     

客户关系管理中的数据挖掘技术的应用探讨

伴随着企业日常运作和客户交流,企业逐渐积累了大量的、复杂的销售数据和客户数据。如何够有效利用这些数据,是当前许多企业所最为关注的焦点。我们将数据挖掘技术引入到客户关系管理中,利用数据挖掘技术对企业销售数据和客户数据进行深入挖掘,找出客户的潜在需求模式和消费模式,并以此指导企业生产运作,提高企业的市场竞争力和客户满意度。     

Equity-premium prediction: Attention is all you need

Predictions of stock returns are greatly improved relative to low-dimensional forecasting regressions when the forecasts are based on the estimated factor of large data sets, also known as the diffusion index (DI) model. However, when applied to text data, DI models do not perform well. This paper shows that by simply using text data in a DI model does not improve equity-premium forecasts over the naive historical-average model, but substantial gains are obtained when one selects the most predictive words before computing the factors and allows the dictionary to be updated over time.     

Forecasting election results by studying brand importance in online news

This study uses the semantic brand score, a novel measure of brand importance in big textual data, to forecast elections based on online news. About 35,000 online news articles were transformed into networks of co-occurring words and analyzed by combining methods and tools from social network analysis and text mining. Forecasts made for four voting events in Italy provided consistent results across different voting systems: a general election, a referendum, and a municipal election in two rounds. This work contributes to the research on electoral forecasting by focusing on predictions based on online big data; it offers new perspectives regarding the textual analysis of online news through a methodology which is relatively fast and easy to apply. This study also suggests the existence of a link between the brand importance of political candidates and parties and electoral results.     

Unit-level linkages between employee commitment to the organization,customer service delivery and customer satisfaction

Service profit chain and service climate research identifies the importance of employee attitudes and employee service behavior as mediating between organizational practices and customer satisfaction. While the importance of employee attitudes and customer service performance are acknowledged, there are calls to more precisely specify proximal mediators between employee attitudes and customer satisfaction. We propose a model in which the relationship between unit-level organizational commitment and customer attitudes is not direct but mediated via employees' customer service delivery including queuing time, serving time and service quality. We conducted a longitudinal unit-level analysis (N = 39) aggregating employee (N over 893) organizational commitment and customer (N over 1248) satisfaction data, and customer service behavior drawn from organizational records. Our model received reasonable support from basic tests of the predictive associations between unit-level organizational commitment, customer-relevant employee behaviors and customer satisfaction; however, organizational commitment was not found to be an important predictor in more rigorous change analyses. The findings as a whole therefore suggest that organizational commitment is a feature of units delivering fast, quality service, but its causal role is as yet unclear.