The rapid rise of digital communication has made emojis and emoticons indispensable visual  cues for expressing concern and intent online. This study applies a pragmatic framework, anchored in relevance theory, to investigate how these symbols shape meaning, tone, and misinterpretation in computer-mediated discourse. A corpus of 1,200 social-media posts (Twitter, and WhatsApp,) is compiled and processed with AntConc, a free corpus-analysis tool. AntConc generated frequency lists, keyword clusters, and concordance lines that divulged the most recurrent emoji/emoticon types and their co-occurring lexical items, facilitating systematic identification of pragmatic functions such as mitigation, irony, and emotional reinforcement. Manual coding of 300 randomly selected instances established that emoticons significantly amplify perceived politeness and reduce vagueness in ambiguous utterances. The findings demonstrate that visual symbols act as relevanceboosting devices, steering readers toward intended interpretations and boosting communicative efficiency. By
incorporating AntConc’s quantitative insights with qualitative pragmatic analysis, this research submits a reproducible methodology for potential digital discourse studies and underlines the importance of visual elements in shaping online interaction.