Please use this identifier to cite or link to this item:

http://hdl.handle.net/10609/65285
Title: Real-Time behavioural stream analysis with Big Data stack technologies
Author: Puertas Ballesteros, Pedro
Director: Andrés Sanz, Humberto
Others: Universitat Oberta de Catalunya
Keywords: Big data
machine learning
business intelligence
Apache Hadoop
wallapop
Issue Date: 30-Jun-2017
Publisher: Universitat Oberta de Catalunya
Abstract: In this report shows the creation of a Big Data solution using the technologies of AWS Kinesis and Spark Streaming and for the visualization part was used Looker. The objective is to be able validate the activity of the users through the action inside the application of Wallapop. To be able to validate the user activity I made a matrix of transitions using the Hidden Markov Model, this model calculate the probability between the actions of the user. Finally, it is to pre-block the user validating the activity and make a decision.
Language: Catalan
URI: http://hdl.handle.net/10609/65285
Appears in Collections:Bachelor thesis, research projects, etc.

Share:
Export:
Files in This Item:
File Description SizeFormat 
Puertas_Ballesteros_Pedro_MEMO¿RIA.docx56.93 MBMicrosoft Word XMLView/Open
Puertas_Ballesteros_Pedro_PRESENTACIO¿.pptx11.27 MBMicrosoft Powerpoint XMLView/Open
ppuertasbTFG0617memoria.pdfMemoria del TFG2.9 MBAdobe PDFView/Open
ppuertasbTFG0617presentación.pdfPresentación del TFG1.07 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons