Paradox Interactive is a Swedish video game developer and publisher which has players all around the world. Paradox’s largest platform in terms of amount of players and revenue is the PC. The goal of this thesis was to make a churn predic-tion model to predict the probability of players churning in order to know which players to focus on in retention campaigns. Since the purpose of churn prediction is to minimize loss due to customers churning the focus was on big-spenders (whales) in Paradox PC games. In order to define which players are big-spenders the spending for players over a 12 month rolling period (from 2016-01-01 until 2018-12-31) was investigated. The players spending more than the 95th-percentile of the total spending for each pe-riod were defined as whales. Defining when a whale has churned, i.e. stopped being a big-spender in Paradox PC games, was done by looking at how many days had passed since the players bought something. A whale has churned if he has not bought anything for the past 28 days. When data had been collected about the whales the data set was prepared for a number of di˙erent supervised machine learning methods. Logistic Regression, L1 Regularized Logistic Regression, Decision Tree and Random Forest were the meth-ods tested. Random Forest performed best in terms of AUC, with AUC = 0.7162. The conclusion is that it seems to be possible to predict the probability of churning for Paradox whales. It might be possible to improve the model further by investi-gating more data and fine tuning the definition of churn. / Paradox Interactive är en svensk videospelutvecklare och utgivare som har spelare över hela världen. Paradox största plattform när det gäller antal spelare och intäk-ter är PC:n. Målet med detta exjobb var att göra en churn-predikterings modell för att förutsäga sannolikheten för att spelare har "churnat" för att veta vilka spelare fokusen ska vara på i retentionskampanjer. Eftersom syftet med churn-prediktering är att minimera förlust på grund av kunderna som "churnar", var fokusen på spelare som spenderar mest pengar (valar) i Paradox PC-spel.För att definiera vilka spelare som är valar undersöktes hur mycket spelarna spenderar under en 12 månaders rullande period (från 2016-01-01 till 2018-12-31). Spelarna som spenderade mer än 95:e percentilen av den totala spenderingen för varje period definierades som valar. För att definiera när en val har "churnat", det vill säga slutat vara en kund som spenderar mycket pengar i Paradox PC-spel, tittade man på hur många dagar som gått sedan spelarna köpte någonting. En val har "churnat" om han inte har köpt något under de senaste 28 dagarna.När data hade varit samlad om valarna var datan förberedd för ett antal olika maskininlärningsmetoder. Logistic Regression, L1 Regularized Logistic Regression, Decision Tree och Random Forest var de metoder som testades. Random Forest var den metoden som gav bäst resultat med avseende på AUC, med AUC = 0, 7162. Slutsatsen är att det verkar vara möjligt att förutsäga sannolikheten att Paradox valar "churnar". Det kan vara möjligt att förbättra modellen ytterligare genom att undersöka mer data och finjustera definitionen av churn.

Retaining high-value and profitable customers is a major strategic objective for many companies. In mature mobile phone markets where growth has slowed, the defection of customers from one network to another has intensified and is strongly fuelled by poor Customer Experience. Trends in the service economy suggest that experience can be exploited as a means of supplying the basis of a new economic offering, ignited in part by the shift that is taking place in the analysis of people’s interaction with digital products. In this light, the research describes a strategic approach to the use of Information Systems as a means of improving Customer Experience. Using Action Research in a mobile telecommunications operator, a Customer Experience Monitoring and Action Response model (CEMAR) is developed that evaluates disparate customer data, residing across many systems, builds experience profiles and suggests appropriate contextual actions where experience is poor. The model provides value in identifying issues, understanding them in the context of the overall Customer Experience (over time) and dealing with them appropriately. The novelty of the approach is the synthesis of data analysis with an enhanced understanding of Customer Experience which is developed implicitly, in real-time and in advance of any instigation by the customer.

Customer retention is an important issue for any business, especially in mature markets such as the UK mobile market where new customers can only be acquired from competitors. Different methods and techniques have been used to investigate customer retention including statistical methods and data mining. However, due to the increasing complexity of the mobile market, the effectiveness of these techniques is questionable. This study proposes Agent-Based Modelling and Simulation (ABMS) as a novel approach to investigate customer retention. ABMS is an emerging means of simulating behaviour and examining behavioural consequences. In outline, agents represent customers and agent relationships represent processes of agent interaction. This study follows the design science paradigm to build and evaluate a generic, reusable, agent-based (CubSim) model to examine the factors affecting customer retention based on data extracted from a UK mobile operator. Based on these data, two data mining models are built to gain a better understanding of the problem domain and to identify the main limitations of data mining. This is followed by two interrelated development cycles: (1) Build the CubSim model, starting with modelling customer interaction with the market, including interaction with the service provider and other competing operators in the market; and (2) Extend the CubSim model by incorporating interaction among customers. The key contribution of this study lies in using ABMS to identify and model the key factors that affect customer retention simultaneously and jointly. In this manner, the CubSim model is better suited to account for the dynamics of customer churn behaviour in the UK mobile market than all other existing models. Another important contribution of this study is that it provides an empirical, actionable insight on customer retention. In particular, and most interestingly, the experimental results show that applying a mixed customer retention strategy targeting both high value customers and customers with a large personal network outperforms the traditional customer retention strategies, which focuses only on the customer‘s value.

<p>The aim of this study is to illustrate, adapt and develop methods of survival analysis in analysing and estimating customer survival time in subscription-based businesses. Two particular objectives are studied. The rst objective is to redene the existing survival analysis techniques in business terms and to discuss their uses in order to understand various issues related to the customer-rm relationship.</p>

This dissertation presents a novel modeling framework for understanding the dynamics of peer-to-peer (P2P) networks under churn (i.e., random user arrival/departure) and designing systems more resilient against node failure. The proposed models are applicable to general distributed systems under a variety of conditions on graph construction and user lifetimes. The foundation of this work is a new churn model that describes user arrival and departure as a superposition of many periodic (renewal) processes. It not only allows general (non-exponential) user lifetime distributions, but also captures heterogeneous behavior of peers. We utilize this model to analyze link dynamics and the ability of the system to stay connected under churn. Our results offers exact computation of user-isolation and graph-partitioning probabilities for any monotone lifetime distribution, including heavy-tailed cases found in real systems. We also propose an age-proportional random-walk algorithm for creating links in unstructured P2P networks that achieves zero isolation probability as system size becomes infinite. We additionally obtain many insightful results on the transient distribution of in-degree, edge arrival process, system size, and lifetimes of live users as simple functions of the aggregate lifetime distribution. The second half of this work studies churn in structured P2P networks that are usually built upon distributed hash tables (DHTs). Users in DHTs maintain two types of neighbor sets: routing tables and successor/leaf sets. The former tables determine link lifetimes and routing performance of the system, while the latter are built for ensuring DHT consistency and connectivity. Our first result in this area proves that robustness of DHTs is mainly determined by zone size of selected neighbors, which leads us to propose a min-zone algorithm that significantly reduces link churn in DHTs. Our second result uses the Chen-Stein method to understand concurrent failures among strongly dependent successor sets of many DHTs and finds an optimal stabilization strategy for keeping Chord connected under churn.

Abstract The market environment of Mobile Phone Industry has had rapid change in the past decade. It is almost same important to gain the loyalty of existing customers as to gain the new customers in the current competitive environment. For many companies in this industry, Customer Relationship Management has been become a significant issue in their strategies. This research is based on the concept of Customer Relationship Management. The purposes of this research are to probe the performance of Customer Retention Programs and the churn reason in mobile phone industry. The research datum were collected from TAT Corp.(TransAsia Telecommunications) including Customer Retention Programs as ¡§Second Honeymoon Program¡¨, ¡§New Second Honeymoon Program¡¨ and ¡§Talking Reward Program¡¨ and a survey from the customers who had been deactivated in February and March of the year 2002. The analysis of the retention programs show significantly relevant on AGE, GENDER, CUSTOMER LEVELS, TENURE, and RATE PLAN factors. The inspection results of survey datum in the churn reasons are explained individually. The suggestions are made after discussing the policies and strategies of TAT Corp.

<p>The aim of this study is to illustrate, adapt and develop methods of survival analysis in analysing and estimating customer survival time in subscription-based businesses. Two particular objectives are studied. The rst objective is to redene the existing survival analysis techniques in business terms and to discuss their uses in order to understand various issues related to the customer-rm relationship.</p>

This research focuses on introducing a novel concept to design a scalable, hierarchical interest-based overlay Peer-to-Peer (P2P) system. We have used Linear Diophantine Equation (LDE) as the mathematical base to realize the architecture. Note that all existing structured approaches use Distributed Hash Tables (DHT) and Secure Hash Algorithm (SHA) to realize their architectures. Use of LDE in designing P2P architecture is a completely new idea; it does not exist in the literature to the best of our knowledge. We have shown how the proposed LDE-based architecture outperforms some of the most well established existing architecture. We have proposed multiple effective data query algorithms considering different circumstances, and their time complexities are bounded by (2+ r/2) only; r is the number of distinct resources. Our alternative lookup scheme needs only constant number of overlay hops and constant number of message exchanges that can outperform DHT-based P2P systems. Moreover, in our architecture, peers are able to possess multiple distinct resources. A convincing solution to handle the problem of churn has been offered. We have shown that our presented approach performs lookup queries efficiently and consistently even in presence of churn. In addition, we have shown that our design is resilient to fault tolerance in the event of peers crashing and leaving. Furthermore, we have proposed two algorithms to response to one of the principal requests of P2P applications’ users, which is to preserve the anonymity and security of the resource requester and the responder while providing the same light-weighted data lookup.

La minería de datos es una nueva tecnología que está cobrando relevancia en la actualidad, su utilidad para resolver complejos problemas a lo que se enfrentan las empresas (de múltiples variables y casos) ha dado entrada a la aplicación e investigación sobre la misma. Sin embargo, esta tecnología no es una heurística cualquiera, se fundamenta en la rama de las ciencias de la computación denominada inteligencia artificial y las matemáticas mediante la estadística. En un comienzo, las empresas sólo se preocupaban por el almacenamiento de los datos, datos históricos que permitían cálculos matemáticos simples con una finalidad, la generación de reportes. De esta manera, se buscaba responder las preguntas referentes al control del negocio. Posteriormente se profundizaron estas preguntas de control hasta llegar a la creación de un repositorio consolidado, expresado en la tecnología de data warehouse. En la actualidad Las exigencias de los consumidores cada día aumentan más, puesto que la competencia comienza a ser más dinámica, por ende, para establecer una ventaja competitiva, las empresas requieren responder preguntas que van más allá de los datos históricos, es decir, necesitan extraer información que pueda ser útil para el futuro, y de esta manera, dejar el paradigma de una empresa reactiva y pasar a ser una entidad proactiva y preventiva. En este nuevo desafío aparece la tecnología de minería de datos, la cual va inserta en un procedimiento Knowledge Discovery on Databases (KDD), puesto que para obtener información del futuro se debe estar seguro del presente. Esta tecnología se aplica actualmente en variadas empresas, sin embargo, no se vislumbra explícitamente. Las personas son afectadas por ella como parte de un paradigma de consumismo, cuando compran un producto y se le hace un descuento, un aviso publicitario mencionando la promoción de un nuevo producto, cuando se les ofrece un crédito bancario o se les llama telefónicamente para mejorar un servicio que ya tienen contratado, e incluso cuando ingresan a Internet para navegar en sus redes sociales o buscar información. También se ve en los avances biológicos como un diagnóstico rápido y efectivo, una cura basada en la ingeniería genética, entre otros. Actualmente la minería de datos se ha subdivido en múltiples ramas según su aplicación, es así, como se pueden encontrar distintos tipos de minería: Web, de Texto, de Procesos. Estos solamente generan la diferencia en la perspectiva en que se ejecuta el KDD, siendo el último tipo el más reciente. Cabe mencionar que los principales algoritmos de han adaptado según su uso y día a día se implementan mejoras sobre los mismos. Análogamente, también, se desarrollan nuevas formas de valorización sobre sus resultados. Esta memoria busca investigar sobre el KDD y las distintas técnicas que pueden ser utilizadas, para luego aplicarlas a un producto particular en una empresa determinada. En ella se describen todos los procesos por los cuales se transcurrió cada uno visto desde el punto de vista del KDD, por lo que su estructura es como realizar un KDD a un documento de esta índole. Sin embargo, no todo fue la aplicación, puesto que se refinan los modelos y algoritmos tanto de transformaciones como de imputaciones de datos, lo que converge en un aprendizaje incremental, en el que cada intento es expresado como relevante puesto que destaca una etapa particular del KDD. Además, de describir la aplicación del KDD se añade una evaluación comercial utilizando recursos de la compañía y bajo el soporte del área de Aseguramiento de Ingresos y la Vicepresidencia Comercial. En base a esta evaluación comercial, se tiene la evaluación técnica de cada modelo y las peculiaridades que se forman al efectuar el contraste entre ambas. Adicionalmente se evalúa monetariamente los resultados obtenidos desde dos puntos de vista, lo que conlleva al establecimiento de propuestas futuras. Agregado a lo anterior, se presentan problemáticas no documentadas, debido a que su acontecer es propio dentro de lo que es desarrollar un proyecto que tiene al KDD como eje articulador. A su vez, se muestran soluciones y planteamientos para ingresar un proyecto a un área determinada, en otras palabras, se presentan herramientas que ayudan a generar confianza al interior de una empresa para que origine un cambio a nivel organizacional respecto a esta tecnología. Finalmente se concluyen los aprendizajes y las acciones correctivas que debiesen ejecutarse en caso de implementar el piloto a nivel operacional.

The iGaming industry has exploded the last decade. With more competitors on the market with low switching costs, retention of players is an issue not yet resolved. Attracting players through offers has made the iGaming industry less lucrative due to its high competition, where a “bonus war” has taken over. This survey aims to find out how iGaming companies can decrease their churn rate through CRM and loyalty programs to increase revenue and thereby creating a long-term relationship with the player. Based on theory of switching costs and loyalty programs, this thesis investigates how different factors influence customer retention. Interviews with four competing casinos were made to see how they are working progressively towards decreased churn rate. The information provided concluded that management sometimes lacked the understanding of how to utilize consumer information and way of communication to increase their revenue. Loyalty programs were used to somewhat extent and correlates partly to what Berman (2006) describes as an important strategy for customer retention in a competitive, homogenous market. Further research is recommended from a legal, ethical and marketing cost perspective. / IGaming-industrin har exploderat det senaste decenniet. Med en marknad med låga bytesbarriärer och fler konkurrerande företag än någonsin är problemet med att behålla kunder ännu inte löst. Att locka spelare genom erbjudanden har gjort iGaming-industrin mindre lukrativ på grund av den höga konkurrensen och ett "bonuskrig" har tagit över. Denna undersökning syftar till att ta reda på hur iGaming-företag kan minska antalet avhoppande spelare genom CRM och lojalitetsprogram för att öka intäkterna och därmed skapa ett långsiktigt förhållande till kunden. Baserat på teorin om byteskostnader och lojalitetsprogram undersöker denna avhandling hur olika faktorer påverkar kundretentionen. Intervjuer med fyra konkurrerande casinon gjordes för att se hur de arbetar gradvis mot en ökad behållningsgrad av kunder. Den information som tillhandahålls leder till slutsatsen att ledningen ibland saknade förståelse för hur man använder konsumentinformation och olika sätt att kommunicera för att öka sina intäkter. Lojalitetsprogram användes i viss utsträckning och överensstämmer delvis med vad Berman (2006) beskriver som en viktig strategi för kundretention på en konkurrenskraftig och homogen marknad. Ytterligare forskning rekommenderas ur ett juridiskt, etiskt och marknadsföringskostnadsperspektiv.