A Survey on Abstractive Summarization Models using Machine Learning and Deep Learning Techniques

Abstractive Summarization models has gained popularity in past few years especially Sequential Models with attention. With the rise in data summarizing data is of utmost importance. Various techniques are provided in this paper for the given task and generate summaries. Encoder and decoder units are explained in proper manner along with special additions to enhance the performance of the model. The utilization of profound learning structures in characteristic language preparation entered another time after the presence of the succession to arrangement models in the ongoing decade. These models square measure primarily supported on a handful of repeated neural networks that connect the input and associated output information in an encoder decoder design. Better outcomes were conceivable by adding the Attention Mechanism to the RNN layers. The Transformer based model outperformed past best in class models in interpretation errands.