Mining Celebrity Endorsement Perceptions Using Varieties of Twitter - - PowerPoint PPT Presentation

▶
mining celebrity endorsement perceptions using varieties
SMART_READER_LITE
LIVE PREVIEW

Mining Celebrity Endorsement Perceptions Using Varieties of Twitter - - PowerPoint PPT Presentation

Sep 13, 2022 705 likes •887 views

Mining Celebrity Endorsement Perceptions Using Varieties of Twitter Account Automated Data Maria Oikonomidou GSC 2019 Computer Science Department - University of Crete Celebrity Endorsement : Marketing strategy whose purpose is to use one or

slide-1
SLIDE 1

Mining Celebrity Endorsement Perceptions Using Varieties of Twitter Account Automated Data

Maria Oikonomidou

GSC 2019

Computer Science Department - University of Crete

slide-2
SLIDE 2

Celebrity Endorsement :

Marketing strategy whose purpose is to use one or multiple celebrities to advertise a specific product or service.

Computer Science Marketing

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

slide-3
SLIDE 3

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

slide-4
SLIDE 4

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

Coffee Company

slide-5
SLIDE 5

Keywords

Big Data Social Media Similarity Analysis Graph Analysis Content Analysis Marketing Strategy

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

slide-6
SLIDE 6

We measure directly the strength of association between a brand and a celebrity.

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

The idea

Our approach represents a low-cost, real-time alternative to traditional survey-bassed elicitation methods. We measure consumers view of fit between pairs of celebrities and brands and validate through survey.

slide-7
SLIDE 7

Dataset

Crawled data between 2016 - 2018

700 million tweets 52 million user accounts

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

Four parts – different data selection

Follow metric: Graph - 1 million

directed follow relations

Content metric: 11G tweets List metric: Weighted Graph - 89 million

relations

Favorite metric: Weighted Graph -

206 thousand relations twAler *

slide-8
SLIDE 8

Similarity Metrics

slide-9
SLIDE 9

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

Follow Similarity Metric

Structural similarity, shape of the Twitter Social graph A set of users follows a brand b, another set

  • f users follows a celebrity c

High intersection = high similarity Jaccard index

slide-10
SLIDE 10

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

List Similarity Metric

Twitter users curate lists of other users​ Similarity between users according to number of lists they are placed together High number of common lists = high similarity

slide-11
SLIDE 11

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

Content Similarity Metric

Most active type of participation Content Comparison between followers of the targeted users Vectorize via TF - IDF Cosine Similarity

slide-12
SLIDE 12

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

Favorite Similarity Metric

"Like" Smallest possible effort Big set of common users favored a pair of targeted users = high similarity Weighted Jaccard Index

slide-13
SLIDE 13

Evaluation

slide-14
SLIDE 14

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

100 celebrities 100 brands

Targeted users Survey Design Evaluation

8 sectors Automobiles,Technology, Retail etc

Association scale 1 to 5 by consumers

Pearson correlation coefficients (r)

Similarity metric results Survey results

slide-15
SLIDE 15

Findings

Automobiles & parts Follow metric Technology Content Metric Favourite metric Financial Services Industrial good List Metric Strong

  • nger Correla

lations

  • ns

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

slide-16
SLIDE 16

Questions?

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

slide-17
SLIDE 17

Maria Oikonomidou mareco@ics.forth.gr GSA 2019

Thank you!