Bibliographic citations
Milla, H., (2017). Propuesta de metodología para mejorar la efectividad de las campañas comerciales de un ISP utilizando Data Mining [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/13263
Milla, H., Propuesta de metodología para mejorar la efectividad de las campañas comerciales de un ISP utilizando Data Mining [Tesis]. : Universidad Nacional de Ingeniería; 2017. http://hdl.handle.net/20.500.14076/13263
@mastersthesis{renati/262627,
title = "Propuesta de metodología para mejorar la efectividad de las campañas comerciales de un ISP utilizando Data Mining",
author = "Milla Caballero, Heber Héctor",
publisher = "Universidad Nacional de Ingeniería",
year = "2017"
}
The companies operating in a timely manner require huge amounts of operational data available. Data mining is a response to this problem, to analyze and automatically exploit large databases to extract useful and not obvious information [1]. One of the major problems of an Internet Service Provider (ISP) is to improve the effectiveness of a marketing campaign by selecting the potencial customer, with a major assurance (better selection). The purpose of this thesis is to solve this problem by proposing a methodology using Data Mining [2]. The thesis is presented with a study and analysis of the process and the tools currently used at typical ISP (TdP) to select potential customers, who are currently using manual processes, they are not using all data available and no there is evidence that they follow a methodology of Data Mining. Then, we study and analyze the data mining methods that exist in the market and, having present that, is an open standard and a comprehensive method; CRISP-DM (Cross Industry Standard Process for Data Mining) is selected as the Data Mining tool to use [3]. With this basis, a proposal to improve the process or methodology for managing ISP campaigns is mode. To consolidate the new proposed methodology campaign management, which includes not used data (Internet traffic applications: P2P, Web, etc.) the new Data base is modeled in UML. On the other hand, is designed in UML part of the proposed process (modelling and deployment), and is implemented in MATLAB using as Data Mining tool the free software WEKA ((Waikato Environment for Knowledge Analysis) Finally the proposed methodology is validated with real data available from a typical ISP (TdP), decision trees as the modeling technique showing that the effectiveness of the campaigns improve.
This item is licensed under a Creative Commons License