An approach to foster agribusiness marketing applying data analysis of social network

• Published in: Computers and electronics in agriculture Vol. 222; p. 109044
• Main Authors: Soares, Nedson D., Braga, Regina, David, José Maria N., Siqueira, Kennya B., Stroele, Victor
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-07-2024
• Subjects: Agribusiness marketing; Cheese market; Dairy derivatives; Data analysis; Social networks; Social networks; Dairy derivatives; Data analysis; Agribusiness marketing; Cheese market
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2024.109044

•Agribusiness increasingly needs smart tools to generate new knowledge.•The processing of large volumes of data using intelligent tools allows the discovery of correlations and the derivation of knowledge in Agribusiness.•This work proposes to foster marketing in Agribusiness, identifying influencers and user communities.•We specify an automated solution for data analysis based on OSN content to extract, process, and recommend communities of influencers through ML algorithms, ontologies and reasoners, complex network metrics, and community detection techniques.
Applying social network data analysis to the agribusiness context can be useful to increase profitability, mainly in the dairy derivatives niche.
The dairy derivatives market needs to recover its profitability. After a 2.9 % GDP growth in 2022,1Source https://www.canalrural.com.br/noticias/pecuaria/leite/leite-confira-como-esta-o-mercado-brasileiro/ (in Portuguese).1 Canal Rural the economic projections indicated only 0.91 % in 2023. Specific strategies to foster this market need to be applied.
To collect information from social networks to find influential people who appreciate dairy derivatives and can influence new potential consumers, we present the IntelDigitalMarketing architecture. Its features encompass social network analysis, recommendations, and context propagation. Through its use, influencers and user communities can be identified who address issues related to specific domains in different social networks, and who can disseminate information to foster specific market niches.
We used the Design Science Research methodology to conduct the study. The solution encompasses techniques such as complex networks, machine learning, and ontologies, to detect market trends. With IntelDigitalMarketing architecture, we processed social network data from X (formerly Twitter), Instagram, and YouTube.
The results showed that the solution can search for communities of digital influencers who talk about dairy derivatives, what they talk about, and the dissemination of information on these social networks.
With the combination of techniques, we can detect new relevant relationships among users that are not detected by other similar solutions. In addition, the proposed solution is online and in real-time, making it easier to follow trends in social networks and with the potential to foster the Agribusiness market.