iosr-JCE  Volume-28 ~ Issue-3 ~ May - Jun. 2026

Abstract: Background : Livestock plays a critical role in the agro-based economy of Bangladesh, yet production is frequently hampered by diseases and inadequate veterinary healthcare services. To overcome these constraints, this study proposes a smart disease diagnostic model that utilizes machine learning (ML) and deep learning (DL) algorithms to process clinical symptoms and field images for early disease detection. Materials and Methods : Using a synthetic clinical dataset of 120 instances and a large-scale image dataset of 8,014 annotated....

Key Word: Machine Learning, Deep Learning, Disease Diagnosis, Farm Animals, YOLOv12, CatBoost, Livestock Management

[1] Department of Livestock Services (DLS), "Livestock Economy at a Glance (2019-2020)," Ministry of Fisheries and Livestock,
Govt. of the People’s Republic of Bangladesh, 2020. [Online]. Available: dls.gov.bd.
[2] L. Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, Oct. 2001, doi: https://doi.org/10.1023/A:101093340 4324.
[3] T. Chen and C. Guestrin, “XGBoost: a Scalable Tree Boosting System,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD ’16, vol. 1, no. 1, pp. 785–794, Aug. 2016, doi: https://doi.org/ 10.1145/2939672.2939785.
[4] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, Sep. 1995, doi: https:// doi.org/10.1007/BF00994018.
[5] V. Garg and K. Garg, “Early Stage Disease Detection Platform in,” vol. 3, no. 11, pp. 13–15, 2016.

Abstract: Abstracts are transformed into Term Frequency Inverse Document Frequency (TF-IDF) vectors and relevance is estimated by cosine similarity against a user query or a target abstract. The method is designed for simple deployment, transparent scoring, and practical use in small to medium collections. Experiments on 5,000 abstracts from arXiv data which is available in Huggingfaces an open online repository of scientific papers, show that TF-IDF with cosine similarity outperforms a keyword overlap baseline. Precision at the top five results improved from 0.265 to 0.327, meaning the proposed method retrieved more relevant papers near the top of the ranked list. The mean reciprocal rank improved from 0.479 to 0.548, meaning the first relevant result appeared earlier in the ranking.

Key Word: Term Frequency-Inverse Document Frequency, Information Retrieval, Abstract Similarity, Cosine Similarity, Content-Based Filtering, Document Ranking

[1]. K. Spärck Jones, "A Statistical Interpretation of Term Specificity and Its Application in Retrieval,".
[2]. G. Salton and C. Buckley, "Term-weighting approaches in automatic text retrieval,"
[3]. C. D. Manning, P. Raghavan, and H. Schütze, Introduction to Information Retrieval.
[4]. S. Robertson, S. Walker, S. Jones, M. Hancock-Beaulieu, and M. Gatford, "Okapi at TREC-3," in Proceedings of the Third Text REtrieval Conference (TREC-3), NIST, 1995.
[5]. Scikit-learn: Machine Learning in Python.
[6]. https://scispace.com/pdf/identifying-the-development-and-application-of-artificial-4pfg5m58k5.pdf.