Top-Up Forecasting of Pre-Paid Mobile Subscribers
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10400.22/20157 |
Resumo: | In an ever-evolving technology world, telecommunications operators must attend to client needs in an effective and speedy manner to strengthen their relationship. The difficulty of this challenge is heightened in Big Data environments where there is a necessity to make sense of the valuable information within data. In the pre-paid telco environment, also known as pay-as-you-go, it is imperative for operators to predict client behaviour efficiently to meet their needs and improve campaigns and notifications, thus improving communication, client retention and revenue. In this dissertation, a novel top-up date and value prediction solution for the prepaid telco environment, is presented. This solution aims to dynamically estimate, for each client, the top-up date and value for the upcoming month. For this, the initial data goes through the developed processing pipeline. The first step is pre-processing, where data is cleaned and transformed. After this, it undergoes a feature engineering and selection step to identify the most relevant features for the prediction of the monthly frequency and value. For the prediction of the targets, several regression techniques were studied both on the offline and online scenario with the help of sliding windows. Using the most efficient technique, the monthly target predictions undergo a processing stage in which they are transformed into the individual top-up date range and top-up monetary value range for the following month. The evaluation of these predicted ranges is based on verifying if the observed event falls within the predicted interval. The solution is implemented in Python and the Jupyter Notebooks environment for data analysis, dimensionality reduction and offline learning experiments. The online learning experiments make use of the Massive Online Analysis (MOA) graphical user interface (GUI) framework. In the end, the designed solution is able to estimate individual top-up activity with an accuracy of approximately 80 % for the date and 70 % for the monetary value. |
id |
RCAP_3eaea8d784975b42bbf5c13ab1deb812 |
---|---|
oai_identifier_str |
oai:recipp.ipp.pt:10400.22/20157 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Top-Up Forecasting of Pre-Paid Mobile SubscribersTelecommunicationsTop-upForecastingBig DataMachine LearningTelecomunicaçõesRecargaPrevisãoIn an ever-evolving technology world, telecommunications operators must attend to client needs in an effective and speedy manner to strengthen their relationship. The difficulty of this challenge is heightened in Big Data environments where there is a necessity to make sense of the valuable information within data. In the pre-paid telco environment, also known as pay-as-you-go, it is imperative for operators to predict client behaviour efficiently to meet their needs and improve campaigns and notifications, thus improving communication, client retention and revenue. In this dissertation, a novel top-up date and value prediction solution for the prepaid telco environment, is presented. This solution aims to dynamically estimate, for each client, the top-up date and value for the upcoming month. For this, the initial data goes through the developed processing pipeline. The first step is pre-processing, where data is cleaned and transformed. After this, it undergoes a feature engineering and selection step to identify the most relevant features for the prediction of the monthly frequency and value. For the prediction of the targets, several regression techniques were studied both on the offline and online scenario with the help of sliding windows. Using the most efficient technique, the monthly target predictions undergo a processing stage in which they are transformed into the individual top-up date range and top-up monetary value range for the following month. The evaluation of these predicted ranges is based on verifying if the observed event falls within the predicted interval. The solution is implemented in Python and the Jupyter Notebooks environment for data analysis, dimensionality reduction and offline learning experiments. The online learning experiments make use of the Massive Online Analysis (MOA) graphical user interface (GUI) framework. In the end, the designed solution is able to estimate individual top-up activity with an accuracy of approximately 80 % for the date and 70 % for the monetary value.Num mundo de tecnologia em constante evolução, as operadoras de telecomunicações devem atender às necessidades dos clientes de forma eficaz e rápida para satisfazêlos. A dificuldade deste desafio é aumentada em ambientes de Big Data, pois surge a necessidade de entender quais são as informações valiosas nos dados. No ambiente de telecomunicações pré-pago, também conhecido como pay as you go, é imperativo que as operadoras prevejam o comportamento de seus clientes de forma eficiente para que suas necessidades sejam atendidas eficientemente e campanhas e notificações aprimoradas possam ser-lhes enviadas com base na sua atividade, melhorando assim a comunicação, retenção de clientes e receita. Nesta dissertação, uma nova solução de previsão de valor e data de recarga, para o ambiente de telecomunicações pré-pago, é apresentada. Esta solução tem como objetivo prever de forma dinâmica a data de recarga e o valor de cada cliente para o mês seguinte. Para isso, os dados iniciais, após serem estudados, são colocados numa pipeline desenvolvida onde a primeira etapa é o pré-processamento onde os dados são limpos e transformados. De seguida, passam por uma etapa de engenharia e seleção de variáveis para obter apenas os variáveis mais relevantes para a previsão dos targets, frequência mensal e valor, respectivamente. Para a previsão dos targets, diversas técnicas de regressão são estudadas tanto no cenário offline como no online com o auxílio de uma janela deslizante. Depois de escolhida a técnica mais eficiente, as previsões mensais previstas passam por uma etapa de processamento na qual são transformadas de modo a obter-se um intervalo de dias e de valor monetário definido para cada cliente para o mês seguinte. A avaliação dessas estimativas é definida com base em averiguar se o evento observado se encontra dentro do intervalo previsto. A solução é implementada utilizando Python no ambiente Jupyter Notebooks para análise de dados, redução de dimensionalidade e experiências de aprendizagem offline. As experiências de aprendizagem online fazem uso da interface gráfica MOA. No final, a solução desenvolvida é capaz de prever a atividade de recarga dos clientes com uma precisão de aproximadamente 80 % para a data e 70 % para o valor.Malheiro, Maria Benedita Campos NevesRepositório Científico do Instituto Politécnico do PortoAlves, Pedro Miguel Ferreira20212024-11-11T00:00:00Z2021-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10400.22/20157TID:202797180enginfo:eu-repo/semantics/embargoedAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-03-13T13:15:04Zoai:recipp.ipp.pt:10400.22/20157Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T17:40:12.948695Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
title |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
spellingShingle |
Top-Up Forecasting of Pre-Paid Mobile Subscribers Alves, Pedro Miguel Ferreira Telecommunications Top-up Forecasting Big Data Machine Learning Telecomunicações Recarga Previsão |
title_short |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
title_full |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
title_fullStr |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
title_full_unstemmed |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
title_sort |
Top-Up Forecasting of Pre-Paid Mobile Subscribers |
author |
Alves, Pedro Miguel Ferreira |
author_facet |
Alves, Pedro Miguel Ferreira |
author_role |
author |
dc.contributor.none.fl_str_mv |
Malheiro, Maria Benedita Campos Neves Repositório Científico do Instituto Politécnico do Porto |
dc.contributor.author.fl_str_mv |
Alves, Pedro Miguel Ferreira |
dc.subject.por.fl_str_mv |
Telecommunications Top-up Forecasting Big Data Machine Learning Telecomunicações Recarga Previsão |
topic |
Telecommunications Top-up Forecasting Big Data Machine Learning Telecomunicações Recarga Previsão |
description |
In an ever-evolving technology world, telecommunications operators must attend to client needs in an effective and speedy manner to strengthen their relationship. The difficulty of this challenge is heightened in Big Data environments where there is a necessity to make sense of the valuable information within data. In the pre-paid telco environment, also known as pay-as-you-go, it is imperative for operators to predict client behaviour efficiently to meet their needs and improve campaigns and notifications, thus improving communication, client retention and revenue. In this dissertation, a novel top-up date and value prediction solution for the prepaid telco environment, is presented. This solution aims to dynamically estimate, for each client, the top-up date and value for the upcoming month. For this, the initial data goes through the developed processing pipeline. The first step is pre-processing, where data is cleaned and transformed. After this, it undergoes a feature engineering and selection step to identify the most relevant features for the prediction of the monthly frequency and value. For the prediction of the targets, several regression techniques were studied both on the offline and online scenario with the help of sliding windows. Using the most efficient technique, the monthly target predictions undergo a processing stage in which they are transformed into the individual top-up date range and top-up monetary value range for the following month. The evaluation of these predicted ranges is based on verifying if the observed event falls within the predicted interval. The solution is implemented in Python and the Jupyter Notebooks environment for data analysis, dimensionality reduction and offline learning experiments. The online learning experiments make use of the Massive Online Analysis (MOA) graphical user interface (GUI) framework. In the end, the designed solution is able to estimate individual top-up activity with an accuracy of approximately 80 % for the date and 70 % for the monetary value. |
publishDate |
2021 |
dc.date.none.fl_str_mv |
2021 2021-01-01T00:00:00Z 2024-11-11T00:00:00Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10400.22/20157 TID:202797180 |
url |
http://hdl.handle.net/10400.22/20157 |
identifier_str_mv |
TID:202797180 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/embargoedAccess |
eu_rights_str_mv |
embargoedAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799131489905934336 |