EUVIMEDEUVIMED entdecken
Crossref

Detection of Online Spread Terrorism using Web Data Mining

Anusha P, Yashaswini J

International Journal of Innovative Research in Computer and Communication Engineering · 25.07.2026 · 14 (07)

DOI: 10.15680/ijircce.2026.1407022 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Innovative Research in Computer and Communication Engineering, online erschienen am 25.07.2026. Verfasst von Anusha P und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The unchecked growth of internet-based platforms has made it far easier for extremist groups to publish propaganda, spread radical ideology, and recruit sympathisers online. Manual monitoring of this content is no longer practical because of the sheer scale and speed at which new web pages appear, and traditional keyword filters tend to miss context, sarcasm, and coded language. This paper presents a lightweight, automated pipeline that combines web mining, Natural Language Processing (NLP) and a Multinomial Naive Bayes classifier to examine the text of a given webpage and estimate how likely it is to contain terrorism-related material. A user simply submits a URL the system fetches the page, strips away scripts, navigation bars and other non-content elements using BeautifulSoup, cleans the remaining text (lower-casing, stop-word removal, tokenisation), converts it into a TF-IDF feature vector, and passes that vector to a Naive Bayes model trained on a labelled dataset of six content categories — Terrorism, Riot, Disaster, Protest, Political and Positive/Safe. The predicted category is then mapped onto a four-tier risk scale (Critical, High, Medium, Low) along with a confidence score, a short list of dominant keywords, and basic page statistics. Testing on a small set of live URLs shows that the pipeline behaves consistently and produces interpretable results, achieving an overall test accuracy of roughly 93% on the held-out dataset. The paper also compares this shallow, resource-efficient approach against heavier deep-learning alternatives reported in recent literature and outlines directions — multilingual support, transformer-based models, and real-time dashboards — for extending the system in future work.

Zitieren

APA

P, A., & J, Y. (2026). Detection of Online Spread Terrorism using Web Data Mining. International Journal of Innovative Research in Computer and Communication Engineering, 14(07). https://doi.org/10.15680/ijircce.2026.1407022

Vancouver

P A, J Y. Detection of Online Spread Terrorism using Web Data Mining. International Journal of Innovative Research in Computer and Communication Engineering. 2026;14(07). doi:10.15680/ijircce.2026.1407022

BibTeX
@article{P2026,
  author = {P, Anusha and J, Yashaswini},
  title = {Detection of Online Spread Terrorism using Web Data Mining},
  journal = {International Journal of Innovative Research in Computer and Communication Engineering},
  year = {2026},
  volume = {14},
  number = {07},
  doi = {10.15680/ijircce.2026.1407022}
}
Kostenlos starten →