Data-driven Approach in Developing a PR Strategy for the HI! Brand

Due to the growing volume of daily generated information in the media environment, the issue of effective use of Big Data processing tools becomes more and more urgent for timely adaptation of the communication and marketing strategies of commercial brands. This study attempts to apply Natural Language Processing algorithms based on machine learning and AI for analysis of large volumes of textual media data in order to identify insights and patterns that may be valuable for further strategic communications planning of the commercial brand in plant-based food industry. The analysis was carried out on three types of collected data - comments in social media platforms, publications in traditional media and reviews on marketplaces. In terms of the study, topic modeling with LDA and text summarization with LLM GPT-4 were carried out. In parallel, the own methodology was implemented based on machine learning algorithms (TextRank, fastText) and basic statistical analysis to identify value motives for consuming plant-based products. Based on the results of the analysis, it became possible to formulate meaningful conclusions and communication recommendations for the plant-based food brand. Keywords: big data analysis, topic modeling, natural language processing, strategic communication planning