Brand Reputation Analysis for Renewable Energy Firms Using Glove Word Embeddings

Brand reputation plays a pivotal role in the growth and sustainability of renewable energy firms, as public trust directly influences adoption and investment. With the surge of digital communication, analyzing reputation through unstructured textual data has become essential. Existing methods often rely on simple sentiment lexicons or frequencybased text mining, which fail to capture semantic nuances, contextual meaning, and domain-specific language used in renewable energy discourse. To address these limitations, the proposed framework employs GloVe word embeddings(GloVe) to transform textual data into semantic vector representations. These embeddings enable robust sentiment classification and topic clustering, improving the detection of subtle reputation signals. The framework is applied to social media and news content to develop a Brand Reputation Index that reflects sentiment, topic prevalence, and engagement patterns. Findings indicate enhanced accuracy in reputation assessment and more precise identification of positive and negative narratives affecting renewable energy firms.

Ҳали таржима қилинмаган