MS / MPhil

Permanent URI for this community

Now showing 1 - 1 of 1

URDU ABSTRACTIVE TEXT SUMMARIZATION USING DEEP LEARNING APPROACHES
(UMT, Lahore, 2024) HAMZA SAEED
One of the most experimental profound learning challenges is text summarization, which reduces the amount of text while retaining its essential and core information. Earlier research in this field will produce impressive outcomes by utilizing highly reliant data. Works about natural language processing is becoming more popular due to implementation issues. In addition to being a key concept in NLP, abstractive text summarization is a popular and challenging topic with variable solutions or outcomes depending on the dataset. The goal of the study is to reformat critical information that has been collected from files, documents, or datasets into a well-simplified text format. The methodology we will use in our study is RNN-based abstractive text summarization. This approach helps the model perform better and is applied to troubleshoot complex issues. One of this paper's features is keyword extraction. Modern technology will be employed to confirm the model's effectiveness. The model's accuracy will be determined and validated using BLEU and ROUGE. The dataset's most prominent feature is how independent each column's relationship is from the others. The paper aims to create condensed text from enormous amounts of data (text format). Sequence-to-sequence and LSTM models are common methods for solving these kinds of problems.