Churn in Social Networks

Karnstedt, Marcel; Hennessy, Tara; Chan, Jeffrey; Basuchowdhuri, Partha; Hayes, Conor; Strufe, Thorsten

doi:10.1007/978-1-4419-7142-5_9

Cited by 27 publications

(31 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Using a Markov network model, the effect of the social network explained 19.5% of the churn and 8.4% of cross-buy (Dierkes et al 2011). Similar results were established in the case of online social network services (Karnstedt et al 2010, Kawale et al 2009). In this direction, Doyle (2007) presented data indicating that in the United States 76% of consumers of certain (more expensive) goods and services consult friends before deciding on the purchase of such items.…”

Section: Figure 1 -Inverse Herfindahl-hirschman Index For Mobile Accesupporting

confidence: 71%

Schumpeterian competition and consumer networks: understanding concentration in the internet access market

Pereira¹

2017

Blucher Engineering Proceedings

View full text Add to dashboard Cite

Markets undergoing fast-paced product evolution impose a significant informational demand on consumers. They may have to decide among relatively heterogeneous offers while lacking the knowledge required for a fully assessed choice of suppliers. In this case, one important information source is the social network wherein the consumer is immersed, by means of the word-of-mouth (WoM). Beyond the pure informational role, we propose that WoM is a possible cause for the significant market concentration observed in at least one prominent hi-tech market, the internet access service, in which the market organization in many countries does not seem entirely explained by the existing economic literature. Applying an innovative approach, based on agent-based simulation and synthetic graph generation, we propose a model that is able to explain the connections between the inter-firm competition and the WoMinfluenced decision process of consumers. The model robustly demonstrates the relevant effects of WoM on the organization of the market in most situations. The results suggest that the competitive configuration of many other real markets could also be significantly influenced by WoM.Keywords: market concentration, consumer behavior, social networks, word-of-mouth, internet access service, agent-based simulation. ResumoMercados em rápida evolução do produto impõem uma demanda informacional significativa sobre os consumidores. Eles podem ter que decidir entre ofertas relativamente heterogêneas, mesmo sem dispor dos conhecimentos necessários para uma escolha totalmente informada dos fornecedores. Neste caso, uma fonte de informação importante é a rede social em que o consumidor está imerso, por meio do "bocaa-boca" (BaB). Para além do papel puramente informativo, propomos que o BaB é uma possível causa da significativa concentração de mercado observada em pelo menos um mercado de alta tecnologia, o serviço de acesso à Internet, no qual a organização do mercado em muitos países não parece inteiramente explicada pelo literatura econômica existente. Aplicando uma abordagem inovadora, baseada na simulação baseada em agentes e na geração de grafos sintéticos, propomos um modelo capaz de explicar as conexões entre a competição entre empresas e o processo decisório dos consumidores, influenciado pelo BaB. O modelo demonstra de forma robusta os efeitos relevantes do BaB na organização do mercado na maioria das situações. Os resultados sugerem que a configuração competitiva de muitos outros mercados reais também pode ser significativamente influenciada pelo BaB.Palavras-chave: concentração de mercado, comportamento do consumidor, redes sociais, boca-a-boca, serviço de acesso à internet, simulação baseada em agentes. Área ABEIN: 1.2 Competição, preços e estruturas de mercado Classificação JEL: C63, D43, L15.

show abstract

Section: Figure 1 -Inverse Herfindahl-hirschman Index For Mobile Accesupporting

confidence: 71%

Schumpeterian competition and consumer networks: understanding concentration in the internet access market

Pereira¹

2017

Blucher Engineering Proceedings

View full text Add to dashboard Cite

show abstract

“…Similar to our work, the work by Karnstedt et al in [1] and [2] examined the prediction of churners from the Irish online community platform Boards.ie, finding that the probability of a user churning was related to the number of prior users with whom the individual has communicated having churned before. The authors examined the social network properties of churners against non-churners (i.e.…”

Section: Related Worksupporting

confidence: 66%

“…Similar to SAP, Server Fault is a platform that is part of the Stack Overflow question answering site collection. 1 The platform functions in a similar vein to SAP by providing users with the means to post questions pertaining to a variety of server-related issues, and allowing other community members to reply with potential answers. 4.…”

Section: Datasetsmentioning

confidence: 99%

Predicting Online Community Churners Using Gaussian Sequences

Rowe

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Knowing which users are likely to churn (i.e. leave) a service enables service providers to offer retention incentives for users to remain. To date, the prediction of churners has been largely performed through the examination of users' social network features; in order to see how churners and non-churners differ. In this paper we examine the social and lexical development of churners and non-churners and find that they exhibit visibly different signals over time. We present a prediction model that mines such development signals using Gaussian Sequences in the form of a joint probability model; under the assumption that the values of churners' and non-churners' social and lexical signals are normally distributed at a given time point. The evaluation of our approach, and its different permutations, demonstrates that we achieve significantly better performance than state of the art baselines for two of the datasets that we tested the approach on.

show abstract

“…Most research on user retention revolved around the wireless telecommunication industry [6,12,15,7,4,5,14,16,18]. Yet, other domains were studied as well, including banking [11,13], grocery retail [2], pay TV [3], online gaming [10], P2P networks [17] and social networks [8,9].…”

Section: Related Workmentioning

confidence: 99%

“…The main approach for churn prediction is to construct a set of features for each user and train a classifier or regressor for the task [6,12,15,7,4,14,18,11,13,2,3,10,17,8,9]. Some features are related to the service and are independent of the user, such as call quality, billing, customer service and pricing [12,15], and may be utilized to model a prior on churn likelihood in the service at a given time.…”

Section: Related Workmentioning

confidence: 99%

Churn prediction in new users of Yahoo! answers

Dror¹,

Pelleg²,

Rokhlenko³

et al. 2012

Proceedings of the 21st International Conference on World Wide Web

105

View full text Add to dashboard Cite

One of the important targets of community-based question answering (CQA) services, such as Yahoo! Answers, Quora and Baidu Zhidao, is to maintain and even increase the number of active answerers, that is the users who provide answers to open questions. The reasoning is that they are the engine behind satisfied askers, which is the overall goal behind CQA. Yet, this task is not an easy one. Indeed, our empirical observation shows that many users provide just one or two answers and then leave.In this work we try to detect answerers that are about to quit, a task known as churn prediction, but unlike prior work, we focus on new users. To address the task of churn prediction in new users, we extract a variety of features to model the behavior of Yahoo! Answers users over the first week of their activity, including personal information, rate of activity, and social interaction with other users. Several classifiers trained on the data show that there is a statistically significant signal for discriminating between users who are likely to churn and those who are not. A detailed feature analysis shows that the two most important signals are the total number of answers given by the user, closely related to the motivation of the user, and attributes related to the amount of recognition given to the user, measured in counts of best answers, thumbs up and positive responses by the asker.

show abstract

Churn in Social Networks

Cited by 27 publications

References 53 publications

Schumpeterian competition and consumer networks: understanding concentration in the internet access market

Schumpeterian competition and consumer networks: understanding concentration in the internet access market

Predicting Online Community Churners Using Gaussian Sequences

Churn prediction in new users of Yahoo! answers

Contact Info

Product

Resources

About