Efficient and Effective Methodologies for Exploring and Prediction Movement Patterns in Large Networks



Al-Zeyadi, Mohammed Ghaiz
(2018) Efficient and Effective Methodologies for Exploring and Prediction Movement Patterns in Large Networks. PhD thesis, University of Liverpool.

[img] Text
201017403_October 2018_2.pdf

Download (10MB)

Abstract

In the era of Big Data the prevalence of networks of all kinds has grown dramatically, and analysing (mining) such networks to support decision-making processes has become an extremely important subject for research, typically with a view to some social and/or economic gain. This thesis describes research work within the theme of Movement Pattern Mining (MPM) as applied to large network data. MPM is a type of frequent pattern mining that provides observation into how information is exchanged between objects in large networks. In the context of the work described in this thesis, the focus is on how the concept of Movement Patterns (MPs) can be extracted from large networks efficiently and effectively, and how such movement patterns can best be utilised so as to predict future movement. The work describes how, by utilising big data facilities like Share/Distribute Memory Systems and Hadoop/MapReduce, novel data mining based techniques can be used, not only to extract MPs from large networks, but also how they can be utilised for prediction purposes. To this end, the works in this thesis are divided into two parts. The first part is concerned with an investigation of an efficient mechanism for MPM. The second part is concerned with the utilisation of the extracted MPs in the context of prediction. For evaluation purposes, two large network datasets were used: The Great Britain Cattle Tracking System database and the Jiayuan Social Network. The evaluation indicates that an efficient and effective mechanism for identifying and extracting MPs form large networks, and subsequently using then MPs for prediction purposes, has been established.

Item Type: Thesis (PhD)
Divisions: Fac of Science & Engineering > School of Electrical Engineering, Electronics and Computer Science
Depositing User: Symplectic Admin
Date Deposited: 28 Nov 2018 14:16
Last Modified: 03 Mar 2021 17:07
DOI: 10.17638/03027360
Supervisors:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3027360