IDENTIFYING ARTIFICIAL NOSTALGIA IN MARKETING CONTENT USING NLP

Abstract

In contemporary marketing, nostalgic appeal has emerged as a key strategy for building emotional connections with audiences. Brands increasingly recreate elements of the past to evoke sentimental feelings, even among individuals who may not have lived through the referenced era. This study introduces a Natural Language Processing (NLP)-based framework designed to automatically detect such “artificial nostalgia” in marketing texts. The proposed model uses TF-IDF vectorization for transforming advertisement content into numerical features and employs the XGBoost algorithm for classification. Synthetic Minority Oversampling Technique (SMOTE) is applied to balance the dataset. The resulting system achieved an accuracy of 84.16%, showing strong potential for recognizing nostalgiadriven marketing language. These findings highlight the relevance of AI methods for analyzing emotional intent in advertisements and assisting marketers in designing data-driven campaigns.