Kannada Text Summarization

Abstract

The internet has caused a humongous growth in the amount of data available to the common man. Summaries of documents can help find the right information and are particularly effective when the document base is very large. There are mainly two methods for text summarization i.e. extractive and abstractive methods. The proposed model used a combination of both extractive and abstractive methods for Kannada text summarization. The algorithm extracts important sentences from Kannada text collected from online resources. The proposed model used a combination of unsupervised LSTM and TF-IDF techniques, which is the two-step summarization method. Evaluation of the model is done using Rouge which yielded 94% accurate results.