Cross-company customer churn prediction in telecommunication: A comparison of data transformation methods

Cross-Company Churn Prediction (CCCP) is a domain of research where one company (target) is lacking enough data and can use data from another company (source) to predict customer churn successfully. To support CCCP, the cross-company data is usually transformed to a set of similar normal distribution of target company data prior to building a CCCP model. However, it is still unclear which data transformation method is most effective in CCCP. Also, the impact of data transformation methods on CCCP model performance using different classifiers have not been comprehensively explored in the telecommunication sector. In this study, we devised a model for CCCP using data transformation methods (i.e., log, z-score, rank and box-cox) and presented not only an extensive comparison to validate the impact of these transformation methods in CCCP, but also evaluated the performance of underlying baseline classifiers (i.e., Naive Bayes (NB), K-Nearest Neighbour (KNN), Gradient Boosted Tree (GBT), Single Rule Induction (SRI) and Deep learner Neural net (DP)) for customer churn prediction in telecommunication sector using the above mentioned data transformation methods. We performed experiments on publicly available datasets related to the telecommunication sector. The results demonstrated that most of the data transformation methods (e.g., log, rank, and box-cox) improve the performance of CCCP significantly. However, the Z-Score data transformation method could not achieve better results as compared to the rest of the data transformation methods in this study. Moreover, it is also investigated that the CCCP model based on NB outperform on transformed data and DP, KNN and GBT performed on the average, while SRI classifier did not show significant results in term of the commonly used evaluation measures (i.e., probability of detection, probability of false alarm, area under the curve and g-mean).     

Acculturation orientations of competing host communities toward valued and devalued immigrants

This study explored how members of competing host communities may endorse different acculturation orientations towards valued and devalued immigrants settled in the bilingual city of Montreal. Francophone (N=422) and Anglophone (N=401) host community students from low-, medium- and high-ethnic diversity colleges completed the host community acculturation scale (HCAS) along with a questionnaire monitoring a range of social psychological correlates of acculturation orientations. Results showed that Anglophones were more integrationist and individualist and less assimilationist than Francophones. Both Francophones and Anglophones were more integrationist and individualist and less assimilationist, segregationist and exclusionist toward valued than devalued immigrants. Attending a multiethnic college had no influence on respondents’ acculturation orientations. The more welcoming acculturation orientations of Anglophones compared to Francophones was corroborated by results obtained with immigrants attending the French and the English colleges. Social psychological profiles of the respective acculturation orientations were similar across both host communities thus supporting the construct validity of the HCAS. Attitude toward co-national outgroup was a significant determinant of each acculturation orientations, suggesting that intergroup relations between rival host communities has an impact on acculturation orientations toward immigrants.     

GM(1,1)模型灰色预测法Visual Foxpro计算程序

本文详细介绍了在Visual Foxpro6．O中编制的MGl1．PRG计算程序及其运行方法。     

中国科学计量指标数据库的设计与实现

简单介绍了中国科学计量指标数据库, 详细阐述了如何进行数据分析和系统设计, 以及指标数据库的功能实现。     

Verteilte Metadatenverwaltung und Anfragebearbeitung für Internet-Datenquellen

Zusammenfassung.   Wir pr?sentieren in dieser Arbeit das ObjectGlobe-System, ein offenes und verteiltes Anfragebearbeitungssystem auf Internet-Datenquellen. ObjectGlobe erweitert die begrenzten F?higkeiten des Internets bezüglich Anfragebearbeitung, indem es einen offenen Marktplatz schafft, in dem eine Vielzahl von Anbietern Rechenzeit, Daten und Anfrageoperatoren bereitstellen. Den Schwerpunkt dieses Beitrags bildet die im Rahmen des ObjectGlobe-Projekts entwickelte Metadatenverwaltung MDV, ein verteiltes Metadaten-Management-System. Es besitzt eine 3-schichtige Architektur und unterstützt Caching und Replikation in der Mittelschicht, so dass Anfragen lokal ausgewertet werden k?nnen. Benutzer spezifizieren die Informationen, die sie ben?tigen und die repliziert werden, mit Hilfe einer besonderen Regelsprache. Um Replikate aktuell zu halten und die Replikation von neuen und relevanten Informationen anzusto?en, verwendet das MDV-System einen neuartigen Publish & Subscribe-Algorithmus. Wir beschreiben diesen Algorithmus im Detail und zeigen, wie er mit Hilfe eines normalen relationalen Datenbanksystems implementiert werden kann. Eingegangen am 19. November 2001 / Angenommen am 30. Juli 2002     

网络资源界面整合和AGENT界面实现探讨

网络资源的多样性和异地性,影响了用户对网络资源的有效利用,文章探讨了网络资源界面整合的策略和实现方法及智能型Agent界面的设计思想和实现方法。     

情感教学在创新教育中的积极作用

创新教育是全民素质提高的基本条件。它强调以促进学生的整体素质的发展为目标，培养学生的创新意识，创新能力与创造性人格。情感教学能有力地促进教育创新，要搞好创新教育，必须使情感教学和创新教育有机地相结合。     

数据挖掘在数字图书馆中的应用

在描述数据挖掘技术与方法基础之上，探讨了数据挖掘在数字图书馆中的应用空间以及其所具有的巨大应用价值。     

论"二分法"的终结与"三分法"的勃兴

二点论比一点论具有更深刻更果断的特点。但由于它难以在动态中把握持久平衡，无法捕捉两极之间的那个“第三者”，因而不能最终战胜形而上学。能够驾驭两极，游刃于“第三者”的只有三点论，它的最初表现形态是由分到合，即由一分为二到合二而一的，一分为三则是它发展及转化的最高环节。所谓一分为三，就是对多样性世界的三分看法，它反映了事物的全面(正面、反面，侧面)，把握了矛盾的全体(静态矛盾、动态矛盾，矛盾中间)，概括了人类文明的全局(物质文明，政治文明、精神文明)，是科学的世界现与方法论。“三”便是万物的基元。     

钻石开放获取模式的特点、潜在问题及解决建议

[目的/意义] 开放获取模式有多种途径,其中"钻石开放获取(DOA)"是一种新兴模式,在保持科学共同体学术独立性方面逐渐受到非营利出版商的重视。[方法/过程] 采用文献调查法、网络调查法以及内容分析法,解读DOA模式的特点、追踪其实践探索,调研DOA平台的发展情况,分析DOA学术交流生态系统的核心要素。[结果/结论] 挖掘DOA模式的潜在问题,包括资金的可持续性不足、多语言科研成果的国际传播障碍、数字存档的永久保存不足等;聚焦解决措施包括经费补充机制、鼓励多语种的科学知识流通、加强DOA资源的永久保存等。提出构建我国开放获取能力中心以及为非营利出版商提供开放获取的新路径,为我国开放获取的理论与实践提供启示。