Comparing a Traditional Approach for Financial Brand Communication Analysis With a Big Data Analytics Technique

Abstract

Although large amounts of data are now available to companies, mere possession of these data is not sufficient, and for better business decisions, it is necessary to perform thorough data analysis. Nowadays, social networks services (SNS) have become important data sources. The rapid growth of SNS has led to their wide use in various research trends in social sciences. In this paper, we aim to enhance the current understanding of the possibilities offered by social data for brand communication analysis in the financial sector. To this end, a traditional methodology and a digital methodology are used to investigate the brand image of the financial entities. The traditional methodology is the Periodic Evaluation of the Image (PEI). The digital methodology is sentiment analysis, a machine learning technique for big data analytics in social sciences using an algorithm developed in Python. The data are analyzed using both methodologies, and then, their results are compared. The findings suggest that while the results obtained using the method based on big data are consistent with the results obtained with the traditional methodology, the former method allows for easier and faster data analysis. The limitations of this paper relate to the size of the sample, the studied sector, and the scope of the reviewed literature.

Keywords

• Big data applications
• machine learning
• sentiment analysis
• social network services
• Twitter