In this context, "deep text" refers to the application of techniques to Natural Language Processing (NLP) .
: Explaining the decision-making process of "black-box" Deep Learning (DL) models used in text classification , particularly within the biomedical domain. 113941
: These models often require large datasets and can be sensitive to "adversarial noise" (small character-level changes that fool the AI). In this context, "deep text" refers to the
: Sentiment analysis of customer reviews, biomedical literature summarization, and disease-treatment classification. Key Details of the Research Authors : Milad
: It addresses the "black-box" problem where complex neural networks provide accurate results but lack transparency, which is critical for high-stakes fields like healthcare. Understanding "Deep Text"
The identifier refers to a specific research article titled "Post-hoc explanation of black-box classifiers using confident itemsets" , published in the journal Expert Systems with Applications (Volume 165, March 2021). Key Details of the Research Authors : Milad Moradi and Matthias Samwald.
: The paper introduces Confident Itemsets Explanation (CIE) , a model-agnostic method that identifies sets of features (words or tokens) that strongly influence a model's prediction.