Svenja Guhr<p>Based on an analysis of <a href="https://fedihum.org/tags/DHd" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>DHd</span></a> conference abstracts, I trace the evolution of <a href="https://fedihum.org/tags/CLS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>CLS</span></a> methods from 2014 to 2025: from omnipresent <a href="https://fedihum.org/tags/NetworkAnalysis" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>NetworkAnalysis</span></a> and <a href="https://fedihum.org/tags/Annotation" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Annotation</span></a>, to the first appearance of <a href="https://fedihum.org/tags/TopicModeling" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>TopicModeling</span></a> and <a href="https://fedihum.org/tags/SentimentAnalysis" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>SentimentAnalysis</span></a>, to <a href="https://fedihum.org/tags/DeepLearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>DeepLearning</span></a> and <a href="https://fedihum.org/tags/GenerativeAI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>GenerativeAI</span></a>.</p>