Influence maximization for viral marketing in online social networks

In this project, various seeds selection algorithms are implemented to select most influential individuals from social networks. The selected individuals are expected to spread out desired marketing message to highest number of receivers through their connections in a social media marketing campaign...

Full description

Bibliographic Details
Main Author: Lu, Yaman
Other Authors: Tang Xueyan
Format: Final Year Project (FYP)
Language:English
Published: 2018
Subjects:
Online Access:http://hdl.handle.net/10356/74045
_version_ 1811676390603358208
author Lu, Yaman
author2 Tang Xueyan
author_facet Tang Xueyan
Lu, Yaman
author_sort Lu, Yaman
collection NTU
description In this project, various seeds selection algorithms are implemented to select most influential individuals from social networks. The selected individuals are expected to spread out desired marketing message to highest number of receivers through their connections in a social media marketing campaign. The process of information diffusion is simulated by independent cascade model and linear threshold model. A software tool is developed as a graphical interface to facilitate the entire process from taking user input to displaying important outputs from programs. The performance of various seed selection algorithms is evaluated based on the expected influence spread of selected seed nodes and time to complete selection. It was concluded that for social networks with low propagation probability, degree discount algorithm is most suitable, whereas for networks with high propagation probability, single discount selection should be applied. Future improvements of this project include implementation of weighted cascade diffusion model and other seed selection algorithms.
first_indexed 2024-10-01T02:20:43Z
format Final Year Project (FYP)
id ntu-10356/74045
institution Nanyang Technological University
language English
last_indexed 2024-10-01T02:20:43Z
publishDate 2018
record_format dspace
spelling ntu-10356/740452023-03-03T20:33:47Z Influence maximization for viral marketing in online social networks Lu, Yaman Tang Xueyan School of Computer Science and Engineering DRNTU::Engineering In this project, various seeds selection algorithms are implemented to select most influential individuals from social networks. The selected individuals are expected to spread out desired marketing message to highest number of receivers through their connections in a social media marketing campaign. The process of information diffusion is simulated by independent cascade model and linear threshold model. A software tool is developed as a graphical interface to facilitate the entire process from taking user input to displaying important outputs from programs. The performance of various seed selection algorithms is evaluated based on the expected influence spread of selected seed nodes and time to complete selection. It was concluded that for social networks with low propagation probability, degree discount algorithm is most suitable, whereas for networks with high propagation probability, single discount selection should be applied. Future improvements of this project include implementation of weighted cascade diffusion model and other seed selection algorithms. Bachelor of Engineering (Computer Science) 2018-04-23T15:26:33Z 2018-04-23T15:26:33Z 2018 Final Year Project (FYP) http://hdl.handle.net/10356/74045 en Nanyang Technological University 35 p. application/pdf
spellingShingle DRNTU::Engineering
Lu, Yaman
Influence maximization for viral marketing in online social networks
title Influence maximization for viral marketing in online social networks
title_full Influence maximization for viral marketing in online social networks
title_fullStr Influence maximization for viral marketing in online social networks
title_full_unstemmed Influence maximization for viral marketing in online social networks
title_short Influence maximization for viral marketing in online social networks
title_sort influence maximization for viral marketing in online social networks
topic DRNTU::Engineering
url http://hdl.handle.net/10356/74045
work_keys_str_mv AT luyaman influencemaximizationforviralmarketinginonlinesocialnetworks