Mining Profitability of Telecommunication Customers Using K-Means Clustering

Data mining is the powerful technique, which can be widely used for discovering the customers’ behaviors as well as customer’s preferences. As a result, it has been widely used in top level companies for evaluating their Customer Relationship Management (CRM) system today. In this study, a new K-means clustering method proposed to evaluate the cluster customers’ profitability in telecommunication industry in Sri Lanka. Furthermore, RFM model mainly used as an input variable for K-means clustering and distortion curve used to identify optimal number of initial clusters. Based on the results, telecommunication customers’ profitability in Sri Lanka mainly categorized into three levels.

Research on Clustering Analysis and Its Application in Customer Data Mining of Enterprise

The paper study improved K-means algorithm and establish indicators to classify customers according to RFM model. Experimental results show that, the new algorithm has good convergence and stability, it has better than single use of FKP algorithms for clustering. Finally the paper study the application of clustering in customer segmentation of mobile communication enterprise. It discusses the basic theory, customer segmentation methods and steps, the customer segmentation model based on consumption behavior psychology, and the segmentation model is successfully applied to the process of marketing decision support.

Customer Segment Prediction on Retail Transactional Data Using K-Means and Markov Model

Retailing is a dynamic business domain where commodities and goods are sold in small quantities directly to the customers.It deals with the end user customers of a supply-chain network and therefore has to accommodate the needs and desires of a large group of customers over varied utilities.The volume and volatility of the business makes it one of the prospectivefields for analytical study and data modeling.This is also why customer segmentation drives a key role in multiple retail business decisions such as marketing budgeting,customer targeting,customized offers,value proposition etc.The segmentation could be on various aspects such as demographics,historic behavior or preferences based on the use cases.In this paper,historic retail transactional data is used to segment the custo-mers using K-Means clustering and the results are utilized to arrive at a transition matrix which is used to predict the cluster movements over the time period using Markov Model algorithm.This helps in calculating the futuristic value a segment or a customer brings to the business.Strategic marketing designs and budgeting can be implemented using these results.The study is specifically useful for large scale marketing in domains such as e-commerce,insurance or retailers to segment,profile and measure the customer lifecycle value over a short period of time.

Clustering-based Demand Response for Intelligent Energy Management in 6G-enabled Smart Grids

As a typical industrial Internet of things(IIOT)service,demand response(DR)is becoming a promising enabler for intelligent energy management in 6 G-enabled smart grid systems,to achieve quick response for supply-demand mismatches.How-ever,existing literatures try to adjust customers’load profiles optimally,instead of electricity overhead,energy consumption patterns of residential appliances,customer satisfaction levels,and energy consumption habits.In this paper,a novel DR method is investigated by mixing the aforementioned factors,where the residential customer cluster is proposed to enhance the performance.Clustering approaches are leveraged to study the electricity consumption habits of various customers by extracting their features and characteristics from historical data.Based on the extracted information,the residential appliances can be scheduled effectively and flexibly.Moreover,we propose and study an efficient optimization framework to obtain the optimal scheduling solution by using clustering and deep learning methods.Extensive simulation experiments are conducted with real-world traces.Numerical results show that the proposed DR method and optimization framework outperform other baseline schemes in terms of the system overhead and peak-to-average ratio(PAR).The impact of various factors on the system utility is further analyzed,which provides useful insights on improving the efficiency of the DR strategy.With the achievement of efficient and intelligent energy management,the proposed method also promotes the realization of China’s carbon peaking and carbon neutrality goals.

Factor and Cluster Analysis as a Tool for Patient Segmentation Applied to Hospital Marketing in Jordan

Hospital marketing is becoming important for the survival and the prosperity of the health service. In addition, it indirectly acts as a formal feedback channel for the customer requirements, preferences, suggestions and complaints. In this work we have undertaken a survey based marketing study for two main objectives: The first being to better understand the patient clusters through k-means clustering and the second to understand customer perception of the different known quality perspectives through factor rotated and unrotated analysis. All of the questionnaires were designed according to international studies. Based on general descriptive statistics, items classified with higher variance but important, are: clean environment, doctors and nurses capabilities, and specialized doctors. Items that are less important with low variance are: food type, lighting and insurance. Also, items classified as more important with low variance are: recommended, no mistakes, and the cost. Using factor analysis rotated and unrotated reduced the variables into five main variables described as: medical aspects, psychological aspects, cost aspects, hospital image and ease of access and procedures. Using k-means clustering, the customers can be clustered into four main clusters with two of them described as general patient with wide variety of interest, serious cases interested in specialized doctors and food, and very serious case with high stress on equipment, no mistakes.

Methodology for Obtaining Electricity-Price Patterns in Customer Response Programs

The main objective of electricity regulators when establishing electricity markets is to decrease the cost of electricity through competition. However, this scenario cannot be achieved without a full participation of the electricity demand by reacting against electricity prices. The aim of this research is to develop tools for helping customers and aggregators to join price and demand response programs, while helping them to hedge against the risk of short-term price volatility. In this way, the capacity of and hybrid methodology （Self-Organizing Maps and Statistical Ward＇s Linkage） to classify high electricity market prices is analysed. Besides, with the help of Non-Parametric Estimation, some price-patterns were found in the abovementioned clusters. The contained knowledge within these patterns supplies customer market-based information on which to base its energy use decisions. The interest for this participation of customers in markets is growing in developed countries to obtain a higher elasticity in demand. Results show the capability of this approach to improve data management and select coherent policies to accomplish cleared demand offers amongst different price scenarios in a more flexible way.

Data Driven Customer Segmentation for Vietnamese SMEs in Big Data Era

Almost Vietnamese big businesses often use outsourcing services to do marketing researches such as analysing and evaluating consumer intention and behaviour,customers’satisfaction,customers’loyalty,market share,market segmentation and some similar marketing studies.One of the most favourite marketing research business in Vietnam is ACNielsen and Vietnam big businesses usually plan and adjust marketing activities based on ACNielsen’s report.Belong to the limitation of budget,Vietnamese small and medium enterprises(SMEs)often do marketing researches by themselves.Among the marketing researches activities in SMEs,customer segmentation is conducted by tools such as Excel,Facebook analytics or only by simple design thinking approach to help save costs.However,these tools are no longer suitable for the age of data information explosion today.This article uses case analysing of the United Kingdom online retailer through clustering algorithm on R package.The result proves clustering method’s superiority in customer segmentation compared to the traditional method(SPSS,Excel,Facebook analytics,design thinking)which Vietnamese SMEs are using.More important,this article helps Vietnamese SMEs understand and apply clustering algorithm on R in customer segmenting on their given data set efficiently.On that basis,Vietnamese SMEs can plan marketing programs and drive their actions as contextualizing and/or personalizing their message to their customers suitably.

基于RFM的聚类算法在零售市场客户细分研究

客户关系管理作为企业管理的重要组成部分,其客户细分功能直接影响着企业营销战略。为了更好地对零售市场进行客户细分,通过应用某英国零售商数据集,基于RFM模型和4种聚类算法,验证基于RFM模型的K-means、DBSCAN、AGNES、GMM等4种聚类算法在UCI Online Retail零售商数据集上的客户分类效果;并利用轮廓系数、卡林斯基-哈拉巴斯指数(CHI)和戴维森堡丁指数(DBI)评价比较上述4种聚类算法的分类结果。实证结果表明:在所选零售商数据集上,K-means和AGNES算法的聚类效果较好,DBSCAN和GMM算法的聚类效果不理想,旨在为机器学习聚类算法在基于RFM模型的客户分类提供参考和借鉴。建议企业重视产出数据,完善企业数据相关制度;结合客户数据特征和企业自身销售特点,有针对性地使用聚类算法进行客户细分,辅助总结客户画像,进而制定有针对性的营销策略。

客户分级优先的即时配送路径规划方法

为了使即时配送企业能以较低的成本提高配送准时性,从而维护并发展高价值客户,首先,针对即时配送客户的特点改进RFM模型,基于已有数据使用DBSCAN算法进行客户聚类,根据聚类结果使用GBDT算法构建客户分级预测模型对即时配送客户进行分级预测。在此基础上,以即时配送的固定成本、变动成本及客户超时点种类、数量为优化目标,构建基于客户分级优先的即时配送路径优化模型,再设计遗传算法对该模型进行求解。最后,以沈阳市某一站式冷链即时配送企业为对象进行实例分析。结果显示,相比该企业原配送方案,应用客户分级优先的即时配送路径规划方法规划后的方案在配送总成本仅提高4.8%的情况下,高价值、潜在高价值客户超时点数量由6减少为2,且超时点均为边缘客户,同时配送总时间减少了7.3%,验证了该方法的有效性。采用该配送路径规划方法,企业的配送成本虽然会小幅增加,但因配送准时性提升,可以更好地维护高价值客户,同时发展潜在高价值客户向高价值客户转变,进而保持或提高长期收益。

工业用户连续参与需求响应的用户基线负荷精准计算方法

提出了一种将K-means聚类分析与长短期记忆神经网络算法结合,通过工业同源组信息进行迁移学习优化的计算方法。该方法实现了对长时间连续参与需求响应的工业用户基线负荷的精准计算,提高了工业用户需求响应效果评价的准确性。通过城市级虚拟电厂平台采集的参与需求响应实践的工业用户电力负荷数据,验证了该方法的有效性。

政务新媒体互动内容分析及诉求回应研究——以人民网留言板数据分析为例

[研究目的]网络问政已成为新时代主要的政民互动方式之一。通过充分挖掘网络政民互动内容的特征,可以了解民众的民意诉求,亦可以及时发现地方政府部门工作中有待改善之处。[研究方法]根据框架理论,采用自然语言处理技术挖掘中部超大城市领导留言板互动内容中的语言特征、内容特征及行为特征,并分析不同特征间的潜在关联。[研究结论]研究表明,语言特征方面,民众留言弱者框架的占比最高,抗争框架所占比例较低。政府偏好使用描述话语和共情话语,规则话语的使用频次较低。内容特征上,各项领域内容特征与城市统计年鉴相应数据存在关联性,能够客观反映各领域存在的问题,且通过聚类能够有效发掘热点议题和实体辅助政府回应。行为特征方面,民众投诉/求助类留言比例与人均地区生产总值具有相关性。政府回复的行为特征与民众留言的内容特征存在潜在关联,其中政府行为的处理类型受民众留言的内容特征影响最大,缺乏关注语言特征和行为特征。

主体-行为-客体语义分析构建技术功效矩阵

针对准确定义技术、功效主题的关键问题,通过分析技术、功效主题在构建技术功效矩阵中的语义共现性,提出一种基于主体-行为-客体语义分析的技术功效矩阵构建方法;基于目标领域制定检索表达式,在国家知识产权局专利数据库中下载相关专利信息数据,并预处理专利数据,得到目标专利信息文档;利用Python语言编程,采用中文分词工具包语言技术平台提取专利信息文档的主体-行为-客体语义结构,结合目标领域语料库、词频-逆文本频率和余弦相似度计算主题词的语义相似度;利用聚类算法Louvain算法实现社区网络划分,以凝练技术、功效主题,并通过主体-行为-客体语义结构的共现关系构建技术功效矩阵;以海底电缆反应力锥切削技术为例,通过专利实例分析验证所提出方法的有效性。结果表明:在分析大量专利实例以构建技术功效矩阵时,所提出的方法可以有效地实现专利实例中主体-行为-客体语义结构的社区网络划分;通过分析社区网络中节点主题的权重确定社区网络主题,提高了主题凝聚的准确性;在海底电缆反应力锥切削技术的专利实例分析中,利用主体-行为-客体语义结构和Louvain算法凝聚了7个技术主题、9个功效主题,验证了所提出方法的有效性。

考虑路网和用户满意度的集群电动汽车主从博弈优化调度策略

传统的电动汽车(electric vehicle,EV)集中优化方法在实际应用中面临调度困难、计算量大、缺乏真实数据支撑等问题,无法准确揭示各主体间的交互行为。为此,提出一种考虑路网和用户满意度的集群EV主从博弈优化调度策略。首先基于真实出行数据和路网数据模拟用户出行行为。其次,负荷聚合商(load aggregator,LA)整合EV负荷资源,对相似出行特性的EV进行聚类。在双层主从博弈模型中,LA作为上层领导者,聚类后的各EV子群作为下层跟随者。考虑EV用户不同消费偏好,通过优化LA定价策略、新能源及储能系统出力计划、EV集群充放电策略实现纳什均衡,达到各主体共赢,并使用改进遗传算法进行求解。最后,利用仿真验证了所提模型可有效提升LA收益及EV用户消费者剩余,增加新能源消纳,并可为不同消费偏好的用户提供差异化服务。

集群中小企业协同新创平台的资源行动机制——构建数字产业集群的路径探索

以往研究较多关注企业借助大型平台赋能推进数字化转型,而对集群中小企业与新创平台的协同共创过程关注不足,在产业集群数字化转型情境下,集群中小企业与新创平台犹如“难兄难弟”,共同面对资源困境。为此,通过对宁波模具集群企业与创元信息模具工业互联网平台共创案例的研究,发现集群中小企业协同新创平台开展数字化转型,是以共同资源焦点驱动的资源行动演进。在资源探索阶段,集群中小企业可以协同新创平台采用“启发式拼凑”,回应“资源盘活”焦点,但难以解决资源池不足的问题;在资源构建阶段,集群中小企业协同新创平台开展“启发式拼凑”和“模仿式编排”的互动,回应“资源匹配”焦点,缓解了资源获取张力,但如何将资源转换成数字化转型能力仍存在诸多挑战;在资源能力化阶段,集群中小企业协同新创平台可以在“启发式拼凑”和“模仿式编排”互动的基础上引入“渐进式协奏”,并开展循环迭代,回应“资源系统”焦点,缓解资源转换张力。基于“相对对称”赋能情境的研究结论对我国加快产业集群数字化转型具有启示意义。

基于AP聚类算法的充电站/光伏电站一体化规划方法

文章基于熵权法构建光伏出力特性指标权重,利用AP聚类算法生成典型场景。同时,计入电动汽车负荷时空分布特性及需求响应能力,建立了负荷联动时空响应模型。基于分时电价与光伏典型场景出力,优化电动汽车充电时序及空间布局,满足充电站距离约束、系统网络约束等前提下,提出了以充电站年总成本最小、用户满意度指标最优的充电站/光伏电站一体化规划方法。最后,通过算例仿真,基于对各个场景下经济性与满意度等指标的权衡考量,求得了充电站/光伏电站一体化规划方案。

基于LRMC模型的B2B定制化生产企业客户细分

客户细分是企业精益管控各类客户的重要手段。研究首先综合考虑B2B定制化企业客户忠诚度、流失趋势、价值贡献和潜在价值等特性构建LRMC(length recency monetary capital)客户细分模型,基于LRMC模型指标需求融合转换企业订单数据和网络爬虫所获取客户特征数据;然后,引入密度峰值思想优化K-means聚类机制,以LRMC参数为输入对客户进行聚类划分,并采用组合赋权法计算不同客户群体加权价值;最后,开展实验,验证所提出模型和方法能更高效地标识客户关键特征并细分客户。提出的客户细分方法可为B2B定制化生产企业精益化客户关系管理提供理论支撑。

面向电动汽车用户的电价套餐模块化设计

合理的电价套餐能够有效地调动电力用户主动参与需求响应的积极性,在减少供电公司的基础建设投资和运行压力的同时,提升售电公司的收益,降低用户的用电费用。针对当前电力市场中用户类型多样,市场竞争压力日增,各方对电价套餐的需求各异,且缺少合适的电价套餐以供选择的情况,提出了一种面向电动汽车(electric vehicle,EV)用户的电价套餐模块化设计方法。首先,通过采用两阶段聚类法生成典型EV用户的充电负荷曲线来确定其充电需求,其中第一阶段使用凝聚层次聚类算法,第二阶段使用K-means聚类方法。其次,运用大规模定制理论并根据典型EV用户的充电需求来设置齐备的电价套餐模块,并计算不同EV用户的个性化充电需求与各模块之间的综合相关度矢量。然后,根据该相关度矢量为每个EV用户分别选择适合的个性化模块组,并逐一配置模块组的相应个性化属性及特征值。最后,结合所选用模块及其属性合特征值配置情况构造考虑需求响应的模块化电价套餐的定价模型。算例结果表明了所提出的电价套餐设计方法是正确、可行的,能够达成供、售电公司以及用户三方共赢的局面。

基于K-means算法的跨国零售商客户细分研究

随着经济全球化及大数据技术的蓬勃发展,跨国零售商之间的竞争日益激烈,根据客户特征进行客户细分,协助客户进行个性化的服务体验,有利于跨国零售商实现精准营销和高效的客户关系管理。为了提高客户细分的精度,本文提出一种基于RFM模型的K-means聚类算法,使用簇内误方差(SSE)和轮廓系数(Silhouette Coefficient)计算聚类个数,优化K值选取。本文选取一家跨国零售商的数据进行实证检验,对细分后的结果进行特征分析,将客户划分为核心型客户、维护型客户和风险型客户三种类别,并为不同客户群体提供差异化营销策略,仅供参考。

基于时间聚类推理的立体车库车位分配策略研究

采用立体车库车辆到达-离去时间数据,通过k-means聚类方法依据不同时段存取车到达频率对车辆进行类别划分,以立方聚类标准为评价指标对划分可信度进行评估。以车辆到达-离去时间划分推理结果及I/O至待存取车位的设备总服务时间与停留时间长短关系,建立立体车库车位分区分配数学模型。定义顾客平均等待时间为立体车库效率评价指标,仿真对比分析就近分配与本文设计聚类推理分区分配的效率指标。仿真结果表明:本文设计的分配策略相较于就近分配策略能有效缩短顾客等待时间,表现为顾客等待时间减少9.5%。研究结果为此类车库车位分配过程提供参考,为提高车库运行效率提供决策支持。

Tamura纹理特征在定制家居饰面分类中的应用

表面装饰是消费者对定制家居的视觉感受性直接影响因素,饰面纹理的研究对于满足消费者审美需求具有重大意义。针对定制家居饰面纹理特征缺少客观分析的现状,将Tamura纹理特征用于定制家居饰面纹理的研究,旨为定制家居设计研发提供参考。依据品牌综合能力及用户熟知程度,搜集了12家国内外饰面厂商所有用于定制家居的饰面图案,共计6 134幅。再根据Tamura纹理特征参数计算方法,利用MatLab软件对其进行特征提取,然后依据手肘法和K均值聚类法进行聚类分析。最终样本被划分为各具特点的五大类,第一类饰面占据了市场主流,第四类饰面市场需求最小。