LDA model tuning for the "Pandemic" subset.

<div><p>Purpose</p><p>The COVID-19 pandemic has changed customer social media engagement behavior, which challenges the establishment of effective marketing strategies to strengthen digital communication with customers and leads to new opportunities for social media competiti...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Xingting Ju (20323838) (author)
منشور في: 2024
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1852024937454239744
author Xingting Ju (20323838)
author_facet Xingting Ju (20323838)
author_role author
dc.creator.none.fl_str_mv Xingting Ju (20323838)
dc.date.none.fl_str_mv 2024-11-25T19:08:05Z
dc.identifier.none.fl_str_mv 10.1371/journal.pone.0313191.g005
dc.relation.none.fl_str_mv https://figshare.com/articles/figure/LDA_model_tuning_for_the_Pandemic_subset_/27902915
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Medicine
Pharmacology
Evolutionary Biology
Developmental Biology
Science Policy
Biological Sciences not elsewhere classified
used amazon comprehend
strengthen digital communication
predicting customer engagement
latent dirichlet allocation
effective marketing strategies
dual focus provides
collected tweets generated
based empirical study
performance metrics show
include &# 8220
customer engagement prediction
brand topic identification
c50 performs best
&# 8220
study presents
brand topics
xlink ">
topics expanded
topic detection
random forest
proposed framework
performing models
particularly strong
new opportunities
framework differentiates
extract sentiments
external factors
comprehensive approach
dc.title.none.fl_str_mv LDA model tuning for the "Pandemic" subset.
dc.type.none.fl_str_mv Image
Figure
info:eu-repo/semantics/publishedVersion
image
description <div><p>Purpose</p><p>The COVID-19 pandemic has changed customer social media engagement behavior, which challenges the establishment of effective marketing strategies to strengthen digital communication with customers and leads to new opportunities for social media competitive intelligence analytics. This study presents a new social media competitive intelligence framework that incorporates not only the detection of brand topics before and during the COVID-19 pandemic but also the prediction of customer engagement.</p><p>Design/Methodology/Approach</p><p>A sector-based empirical study is conducted to illustrate the implementation of the proposed framework. We collected tweets generated by 23 leading American catering brands before and during the pandemic. First, we used Amazon Comprehend and Latent Dirichlet allocation (LDA) to extract sentiments and topics behind unstructured text data. Second, we trained and compared the performance of six machine learning algorithms to find the optimal classifiers.</p><p>Findings</p><p>The study reveals significant shifts in social media engagement topics following the COVID-19 pandemic. Pre-pandemic topics primarily included “Food and lifestyle”, “Promotion”, “Food ordering”, “Food time”, and “Food delivery”. During the pandemic, the topics expanded to include “Social responsibility” and “Contactless ordering”. For predicting customer engagement, the performance metrics show that Random Forest and C5.0 (C50) are generally the best-performing models, with Random Forest being particularly strong for "Likes" and “Retweets”, while C50 performs best for “Replies”.</p><p>Originality</p><p>This framework differentiates itself from existing competitive intelligence frameworks by integrating the influence of external factors, such as the COVID-19 pandemic, and expanding the analysis from topic detection to customer engagement prediction. This dual focus provides a more comprehensive approach to social media competitive intelligence.</p></div>
eu_rights_str_mv openAccess
id Manara_fcff87bdf895ff20ea48fe61914d25a2
identifier_str_mv 10.1371/journal.pone.0313191.g005
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/27902915
publishDate 2024
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling LDA model tuning for the "Pandemic" subset.Xingting Ju (20323838)MedicinePharmacologyEvolutionary BiologyDevelopmental BiologyScience PolicyBiological Sciences not elsewhere classifiedused amazon comprehendstrengthen digital communicationpredicting customer engagementlatent dirichlet allocationeffective marketing strategiesdual focus providescollected tweets generatedbased empirical studyperformance metrics showinclude &# 8220customer engagement predictionbrand topic identificationc50 performs best&# 8220study presentsbrand topicsxlink ">topics expandedtopic detectionrandom forestproposed frameworkperforming modelsparticularly strongnew opportunitiesframework differentiatesextract sentimentsexternal factorscomprehensive approach<div><p>Purpose</p><p>The COVID-19 pandemic has changed customer social media engagement behavior, which challenges the establishment of effective marketing strategies to strengthen digital communication with customers and leads to new opportunities for social media competitive intelligence analytics. This study presents a new social media competitive intelligence framework that incorporates not only the detection of brand topics before and during the COVID-19 pandemic but also the prediction of customer engagement.</p><p>Design/Methodology/Approach</p><p>A sector-based empirical study is conducted to illustrate the implementation of the proposed framework. We collected tweets generated by 23 leading American catering brands before and during the pandemic. First, we used Amazon Comprehend and Latent Dirichlet allocation (LDA) to extract sentiments and topics behind unstructured text data. Second, we trained and compared the performance of six machine learning algorithms to find the optimal classifiers.</p><p>Findings</p><p>The study reveals significant shifts in social media engagement topics following the COVID-19 pandemic. Pre-pandemic topics primarily included “Food and lifestyle”, “Promotion”, “Food ordering”, “Food time”, and “Food delivery”. During the pandemic, the topics expanded to include “Social responsibility” and “Contactless ordering”. For predicting customer engagement, the performance metrics show that Random Forest and C5.0 (C50) are generally the best-performing models, with Random Forest being particularly strong for "Likes" and “Retweets”, while C50 performs best for “Replies”.</p><p>Originality</p><p>This framework differentiates itself from existing competitive intelligence frameworks by integrating the influence of external factors, such as the COVID-19 pandemic, and expanding the analysis from topic detection to customer engagement prediction. This dual focus provides a more comprehensive approach to social media competitive intelligence.</p></div>2024-11-25T19:08:05ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pone.0313191.g005https://figshare.com/articles/figure/LDA_model_tuning_for_the_Pandemic_subset_/27902915CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/279029152024-11-25T19:08:05Z
spellingShingle LDA model tuning for the "Pandemic" subset.
Xingting Ju (20323838)
Medicine
Pharmacology
Evolutionary Biology
Developmental Biology
Science Policy
Biological Sciences not elsewhere classified
used amazon comprehend
strengthen digital communication
predicting customer engagement
latent dirichlet allocation
effective marketing strategies
dual focus provides
collected tweets generated
based empirical study
performance metrics show
include &# 8220
customer engagement prediction
brand topic identification
c50 performs best
&# 8220
study presents
brand topics
xlink ">
topics expanded
topic detection
random forest
proposed framework
performing models
particularly strong
new opportunities
framework differentiates
extract sentiments
external factors
comprehensive approach
status_str publishedVersion
title LDA model tuning for the "Pandemic" subset.
title_full LDA model tuning for the "Pandemic" subset.
title_fullStr LDA model tuning for the "Pandemic" subset.
title_full_unstemmed LDA model tuning for the "Pandemic" subset.
title_short LDA model tuning for the "Pandemic" subset.
title_sort LDA model tuning for the "Pandemic" subset.
topic Medicine
Pharmacology
Evolutionary Biology
Developmental Biology
Science Policy
Biological Sciences not elsewhere classified
used amazon comprehend
strengthen digital communication
predicting customer engagement
latent dirichlet allocation
effective marketing strategies
dual focus provides
collected tweets generated
based empirical study
performance metrics show
include &# 8220
customer engagement prediction
brand topic identification
c50 performs best
&# 8220
study presents
brand topics
xlink ">
topics expanded
topic detection
random forest
proposed framework
performing models
particularly strong
new opportunities
framework differentiates
extract sentiments
external factors
comprehensive approach