Defining content marketing and its influence on online user behavior: a data-driven prescriptive analytics method

• Tipo: Artigo em Revista Científica Internacional
• Data: 2023

Autores

Barbosa, B (Autor) FEP; José Ramón Saura (Autor) Outra; Zekan, SB (Autor) Outra; Ribeiro-Soriano, D (Autor) Outra

Revista

Título: Annals of Operations Research

ISSN: 0254-5330

Editora: Springer Nature

Indexação

ISI Web of Knowledge - 5 Citações

Scopus

Outras Informações

ID Authenticus: P-00Y-486

DOI: 10.1007/s10479-023-05261-1

Abstract (EN): Content marketing involves producing and distributing content effectively and initially through digital channels. However, digital marketing strategies and business models can succeed only if content marketing is developed correctly. This study aims to develop a relevant theoretical framework linked to content marketing and identify the leading techniques and uses linked to its development. In this context, we developed an innovative data-driven methodology consisting of three steps. In the first phase, sentiment analysis that works with machine learning was conducted with Textblob, and four experiments were performed using support vector classifier, multinomial naive bayes, logistic regression, and random forest classifier. First, we aimed to increase the accuracy of sentiment analysis (negative, neutral and positive) of a sample of user-generated content collected from the social network Twitter. Second, a mathematical topic-modelling algorithm known as latent dirichlet allocation was used to divide the database into topics. Finally, a textual analysis was developed using the Python programming language. Based on the results, we identified 11 topics, of which four were positive (Smart Content, Video Marketing, Podcast, and Influencer Marketing). Six of them were neutral (Content Personalization, Social Media Posts, Blogging, search engine optimization, Advergames, and NFTs), and one was negative (Email Marketing). Our results suggest that companies should use content personalisation ethically, mainly when AI-based techniques are used to predict user behaviors. While content marketing strategies are a fundamental part of digital marketing tactics, they can elicit changes in user online behavior when Big Data or AI algorithms are used. This fact raises concerns about the non-ethical design of online strategies in digital environments and the imperative that content marketing strategies should not be developed with purely economic and profitability interests.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 26