A robust pipelined granular approach for enhanced efficiency in document text summarisation using hierarchical categorisation and feature extraction techniques

The exponential growth of digital textual data presents challenges for traditional text summarisation methods, especially due to natural language complexity and high dimensionality. This work proposes a pipelined granular approach to enhance summarisation efficiency by addressing semantic ambiguity, high dimensionality, and scalability. The methodology comprises two phases: optimised feature selection and advanced rule-based categorisation. In the first phase, a granular approach using a discernibility matrix and iterative optimisation reduces dimensionality while retaining key features for accuracy. The second phase employs hierarchical categorisation to generate precise rules, ensuring adaptability and scalability across diverse domains. Experimental benchmarking and complexity analysis validate the approach, demonstrating notable improvements in summarisation accuracy and computational efficiency. The integration of granular feature extraction and hierarchical categorisation effectively tackles key challenges, making the approach suitable for large-scale document analytics across various applications.