EUVIMEDEUVIMED entdecken
Crossref

Development of a Sentiment Analysis AAUA Newsletter Using Deep Learning Technique

E.O. Aliyu

Advances in Multidisciplinary & Scientific Research Journal Publication · 30.06.2026 · 12 (2) · 59-66

DOI: 10.22624/aims/v12n2p5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Multidisciplinary & Scientific Research Journal Publication, online erschienen am 30.06.2026. Verfasst von E.O. Aliyu.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The identification and extraction of subjective information from textual data is vital for refining institutional communication strategies and improving decision-making processes. While prior studies have concentrated on social media and movie reviews, institutional newsletters, especially within the Nigerian educational context remain largely unexplored. This paper presents the development of a Sentiment Analysis of Adekunle Ajasin University, Akungba-Akoko (SAAAUA) newsletter using a deep learning approach. Twenty-two newsletters published between 2022 and 2024 were collected. Text was extracted using PyMuPDF for digital PDFs and EasyOCR for scanned documents, segmented into paragraphs and refined through noise removal and length normalization. A semi-automatic labelling pipeline produced 196 neutral, 62 positive and 16 negative samples. The BERT-base-uncased model was fine-tuned with a dense classification layer and softmax output, trained using the AdamW optimizer with class-weighted cross-entropy loss and early stopping. The model demonstrated exceptional performance on the Neutral class with (97% precision and recall) and moderate performance on the underrepresented negative class. These findings demonstrate the feasibility of applying transformer-based models to niche institutional corpora and establish a baseline for future natural language processing (NLP) research in Nigerian university communication. Keywords: Sentiment Analysis, BERT, Deep Learning, Natural Language Processing, Institutional Communication, AAUA Newsletter, Text Classification. Aims Research Journal Reference Format: Ayanniyi, F.A. & Omoyajowo, B.S. (PhD) (2026): Development of a Sentiment Analysis AAUA Newsletter Using Deep Learning Technique. Advances in Multidisciplinary Research Journal. Vol. 12 No. 2, Pp 59-66. www.isteams.net/aimsjournal. dx.doi.org/10.22624/AIMS/V12N2P5

Zitieren

APA

Aliyu, E. (2026). Development of a Sentiment Analysis AAUA Newsletter Using Deep Learning Technique. Advances in Multidisciplinary & Scientific Research Journal Publication, 12(2), 59-66. https://doi.org/10.22624/aims/v12n2p5

Vancouver

Aliyu E. Development of a Sentiment Analysis AAUA Newsletter Using Deep Learning Technique. Advances in Multidisciplinary & Scientific Research Journal Publication. 2026;12(2):59-66. doi:10.22624/aims/v12n2p5

BibTeX
@article{Aliyu2026,
  author = {Aliyu, E.O.},
  title = {Development of a Sentiment Analysis AAUA Newsletter Using Deep Learning Technique},
  journal = {Advances in Multidisciplinary & Scientific Research Journal Publication},
  year = {2026},
  volume = {12},
  number = {2},
  pages = {59--66},
  doi = {10.22624/aims/v12n2p5}
}
Kostenlos starten →