"ADHD_Detection_Dataset"

"This dataset comprising posts related to ADHD sourced from Reddit has been utilized. However, certain sentences in the dataset were not relevant to ADHD. Therefore, the author fine-tunes the dataset by extracting sentences containing symptoms associated with ADHD. A predefined list of ADHD-related keywords, representing common symptoms and behaviours such as \"inattentive,\" \"impulsive,\" and \"hyperactive\u201d, was curated to serve as the initial lexicon. To incorporate variations in language and terminology, the lexicon was expanded using WordNet, a lexical database from the \u2018nltk\u2019 library. For each keyword, synonyms were generated and added to the list, ensuring comprehensive coverage of symptom-related expressions. A filtering function is developed to identify the posts containing any of the ADHD keywords or their synonyms. Algorithm 1 defines the step-by-step construction of the fine-tuned ADHD dataset. To complement this dataset, a non-ADHD dataset such as Sentiment 140 has also been incorporated. By integrating ADHD-specific posts with general sentiment data, the proposed model is better equipped to generalize across different types of text, consequenty improving its robustness and performance to identify relevant behavioral and linguistic patterns. "