EUVIMEDEUVIMED entdecken
Crossref

Adaptive Trans‐Residual Long Short Term Memory Model for Automatic Text Summarization in Telugu

M. Varaprasad Rao, Kunal Chakma, Anupam Jamatia, Dwijen Rudrapal

Computational Intelligence · 2026-06 · 42 (3)

DOI: 10.1111/coin.70233 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computational Intelligence, online erschienen am 23.04.2026. Verfasst von M. Varaprasad Rao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern text processing requires large volumes of text to clarify issues, resulting in paragraphs with useless or repetitive information. Effective text summarization helps to extract relevant information and eliminates redundant content. Traditional text summarization models sometimes fail to retain original keywords, particularly when working on large datasets. Therefore an automatic hybrid text summarization framework is proposed to address the challenges by integrating a deep learning with meta‐heuristic algorithms “adaptive trans‐residual long short‐term memory” (ATR‐LSTM), fine‐tuned using an “enhanced serval optimization algorithm” (ESOA). This framework was evaluated on Telugu News NLP, Telugu Books, and TeSum datasets, produced significantly improved summaries. On the TeSum dataset, the model achieved ROUGE and BLEU scores of 92% and 89%, respectively, demonstrating strong summarization performance. In addition to this, the processing time was reduced by 30% compared to the baseline SOA‐ATR‐LSTM, by retaining the original content about 95% to produce cohesive, succinct, and high‐quality summaries, which emphasized both efficiency and effectiveness. The results demonstrate that the proposed method is effective for Telugu Text Summarization and a choice for scaling to other Dravidian languages' text summarization.

Zitieren

APA

Rao, M. V., Chakma, K., Jamatia, A., & Rudrapal, D. (2026). Adaptive Trans‐Residual Long Short Term Memory Model for Automatic Text Summarization in Telugu. Computational Intelligence, 42(3). https://doi.org/10.1111/coin.70233

Vancouver

Rao MV, Chakma K, Jamatia A, Rudrapal D. Adaptive Trans‐Residual Long Short Term Memory Model for Automatic Text Summarization in Telugu. Computational Intelligence. 2026;42(3). doi:10.1111/coin.70233

BibTeX
@article{Rao2026,
  author = {Rao, M. Varaprasad and Chakma, Kunal and Jamatia, Anupam and Rudrapal, Dwijen},
  title = {Adaptive Trans‐Residual Long Short Term Memory Model for Automatic Text Summarization in Telugu},
  journal = {Computational Intelligence},
  year = {2026},
  volume = {42},
  number = {3},
  doi = {10.1111/coin.70233}
}
Kostenlos starten →