<![CDATA[The rapid growth of social media has encouraged teenagers to express their emotions through text-based social media posts. However, existing systems still face limitations in understanding emotional meaning and providing appropriate responses. This study aims to develop a Natural Language Processing (NLP)-based system for detecting emotions in Indonesian-language text and generating contextual motivational responses. The research methodology employed the Cross Industry Standard Process for Data Mining (CRISP-DM), which consists of business understanding, data understanding, data preparation, modeling, evaluation, and deployment stages. Text processing was conducted through tokenizing, filtering, stemming using the Enhanced Confix Stripping algorithm, negation checking, and matching with an emotion lexicon based on Plutchik's Wheel of Emotions classification. The dataset consisted of 14,182 emotion lexicon entries used as reference data and 100 emotional expression sentences collected from 100 respondents as testing data. Evaluation using the Weighted F1-score produced a result of 87%. These findings indicate that the proposed system is capable of identifying emotions and generating relevant motivational responses. The integration of emotion detection and response generation within a single system enables outputs that are adaptive to users’ emotional conditions. However, the system still has limitations in detecting non-standard language and slang expressions, indicating that further lexicon enrichment and more adaptive methods are still required.]]>
