This Chapter provides a detailed overview of neural network and deep learning methods for emotion detection, showcasing the models’ capabilities and complexities in interpreting emotions from linguistic data. The human brain architecture and cognitive abilities have inspired the neural network’s structure. Neural networks learn and classify textual data using neurons (nodes), layers, weights, and biases. We briefly describe the computational processes of the neural networks and the use of attention mechanisms to enhance the models’ accuracy. We then focus on specific deep learning models for detecting emotion categories and dimensions in textual resources, reviewing various methodologies and recent findings, and illustrating deep learning’s effectiveness in detecting emotions in different contexts. We describe the advantages and limitations of deep learning methods in detecting and classifying emotion categories and dimensions. We conclude the Chapter by exploring the significance of transfer and generative learning in emotion detection, highlighting their solutions to data scarcity and ability to interpret natural language.