[Paper Review] Decision Making For Celebrity Branding: An Opinion Mining Approach Based On Polarity And Sentiment Analysis Using Twitter Consumer-Generated Content (CGC)

Ali Nikseresht, Mohammad Hosein Raeisi+1 more|arXiv (Cornell University)|Sep 26, 2021

Digital Marketing and Social Media4 citations

TL;DR

This study proposes a sentiment analysis framework using Twitter consumer-generated content to guide celebrity branding decisions, comparing lexicon-based (Naïve Bayes) and machine learning (Naïve Bayes) approaches. It finds that the machine learning method yields higher accuracy in sentiment classification, enabling data-driven selection of more effective celebrity endorsers for marketing campaigns.

ABSTRACT

The volume of discussions concerning brands within social media provides digital marketers with great opportunities for tracking and analyzing the feelings and views of consumers toward brands, products, influencers, services, and ad campaigns in CGC. The present study aims to assess and compare the performance of firms and celebrities (i.e., influencers that with the experience of being in an ad campaign of those companies) with the automated sentiment analysis that was employed for CGC at social media while exploring the feeling of the consumers toward them to observe which influencer (of two for each company) had a closer effect with the corresponding corporation on consumer minds. For this purpose, several consumer tweets from the pages of brands and influencers were utilized to make a comparison of machine learning and lexicon-based approaches to the sentiment analysis through the Naive algorithm (lexicon-based) and Naive Bayes algorithm (machine learning method) and obtain the desired results to assess the campaigns. The findings suggested that the approaches were dissimilar in terms of accuracy; the machine learning method yielded higher accuracy. Finally, the results showed which influencer was more appropriate according to their existence in previous campaigns and helped choose the right influencer in the future for our company and have a better, more appropriate, and more efficient ad campaign subsequently. It is required to conduct further studies on the accuracy improvement of the sentiment classification. This approach should be employed for other social media CGC types. The results revealed decision-making for which sentiment analysis methods are the best approaches for the analysis of social media. It was also found that companies should be aware of their consumers' sentiments and choose the right person every time they think of a campaign.

Motivation & Objective

• To evaluate and compare the performance of lexicon-based and machine learning approaches in sentiment analysis for celebrity branding decisions.
• To identify which celebrity influencer has a stronger psychological and emotional impact on consumers relative to the brand, based on sentiment in social media content.
• To provide actionable insights for marketers on selecting optimal influencers for future advertising campaigns using automated sentiment and polarity analysis.
• To assess the effectiveness of social media mining (SMM) and big data analytics in enhancing decision-making for market research and brand strategy.

Proposed method

• Collected consumer-generated content (CGC) from Twitter posts related to brands and their celebrity endorsers.
• Applied two sentiment analysis techniques: a lexicon-based approach (Naïve algorithm) and a machine learning approach (Naïve Bayes algorithm).
• Conducted polarity and sentiment analysis on tweets to classify consumer sentiment as positive, negative, or neutral toward both brands and their celebrity endorsers.
• Used WordCloud visualization to identify key sentiment-related terms and themes in the CGC.
• Compared the accuracy and performance of both sentiment analysis methods using quantitative evaluation metrics (implied by accuracy comparison).
• Integrated findings into a decision-making model to guide future influencer selection and campaign optimization.

Experimental results

Research questions

• RQ1 Which sentiment analysis method—lexicon-based or machine learning—yields higher accuracy in classifying consumer sentiment toward brands and celebrities on Twitter?
• RQ2 How do consumer sentiments toward a brand compare with those toward its celebrity endorser, and which entity elicits a stronger emotional response?
• RQ3 Which celebrity influencer (from two per brand) has a more significant psychological impact on consumers, as indicated by sentiment patterns in CGC?
• RQ4 To what extent can sentiment analysis of social media content inform strategic decisions in celebrity branding and market research?

Key findings

• The machine learning-based sentiment analysis (Naïve Bayes algorithm) achieved higher accuracy than the lexicon-based approach in classifying consumer sentiment in Twitter CGC.
• Consumer sentiment toward celebrity endorsers was found to be a significant predictor of brand perception and campaign effectiveness.
• The study identified specific influencers who had a stronger emotional resonance with audiences, enabling data-driven selection for future campaigns.
• The integration of sentiment analysis with social media mining provided actionable insights that can optimize marketing strategy, reduce campaign risk, and improve resource allocation.
• The final decision-making model demonstrated that sentiment-driven insights enhance competitive positioning and strategic planning in market research.
• The results suggest that companies should continuously monitor consumer sentiment and use automated sentiment analysis to select the most effective influencers for branding initiatives.