Disclosed is a system 101 and method 500 for document categorization. The system and method comprising a processor 201 coupled with memory 202 configured to convert input text 301 into bit arrays 302, Latent Semantic Indexing (LSI) technique 203 may be used, which converts semantically similar sentences into similar bit arrays 302. These bit arrays 302 are fed to the Spatial Pooler 204 where it simulates the working of neurons in the brain and gives SDR 303 as the output. The active bits in the SDR 303 represent the neurons which get activated in the Spatial Pooler 204. Since semantically similar text belong to the same category, it is easy to classify the input text 301 into different categories 304.
展开▼
