<![CDATA[Automated text classification involves applying artificial intelligence algorithms to classify text documents into predefined categories. Hence, developing a high-accuracy text categorization model is a significant task, especially in unstructured narratives such as research papers, medical documents, and news articles. This study examines the application of an artificial neural network (ANN) algorithm for categorizing formal and slang English language with the capabilities of popular deep learning frameworks such as TensorFlow and Keras. First of all, the dataset's features were examined through exploratory data analysis (EDA) methods to enhance understanding. Furthermore, the study emphasizes the use of several preprocessing techniques to address the challenge presented by the informal writing style. In addition, adding a list of common English abbreviations greatly improved the accuracy and effectiveness of classifying text. Lastly, the work involves using multiple hyperparameter optimization approaches for further enhancement. The proposed techniques effectively mitigated the impact of heterogeneous and noisy data in both formal and informal language by achieving an improvement of approximately 10% in overall classification accuracy. Additionally, the study contributes to an advancement in the field of text mining and offers practical guidance for optimizing deep learning models in the domain of English text categorization.]]>
Copyrights © 2026
