An approach to summarize multilingual news using deep learning technique
Sumita Gupta, Sapna Gambhir, Rana Majumdar · IET conference proceedings. · 2025
Every single day the volume of data is growing exponentially high aiding more than half of the overall digital content. As now technologies are getting advanced and computing infrastructure is affordable, the digital content is catastrophically growing. The most critical challenge is to distinguish between the structured and unstructured data available in the form of text, image, audio, or videos etc on the internet where most of the content are unstructured. In this paper, deep learning algorithms are used for training and testing the data. Also, news translation and summarization from Hindi website to English is done for those people who are fond of news on the web and are not interested to go through the entire news instead they are looking for highlights or summary of those news. So, the automatic summarization of news articles is required. Here, an automatic text summarizer using extractive summarization approach is proposed and implemented by considering deep learning for categorization of news content and text rank algorithm for summarization. To evaluate accuracy, F-measure and recall of the produced summary, deep learning algorithms are applied. The result produced 99.4% of accuracy using LSTM over other deep learning like CNN, LSTM, CNN-LSTM, BI-LSTM etc.