Heart Disease Prediction Using Machine Learning and Data Mining Techniques
Miss. Sandhya Prajapati
, Mr. Manjunath Gowda , Mrs. Sherilyn Kevin , Dr. Santosh Kumar Singh
cardiovascular disease prediction, heart disease detection, Feature Tokenizer Transformer (FT-Transformer), Gradient Boosted Decision Trees (GBDT), Hybrid machine learning, Data mining techniques, Early diagnosis, Preventive cardiology.
Cardiovascular disease remains a leading global health concern, necessitating the development of robust early detection systems. The advent of large-scale medical datasets presents both an opportunity and a challenge for predictive modeling. This research addresses the complexities of handling a substantial dataset of over 5 million patient records for accurate heart disease prediction. We propose and evaluate a high-performance approach utilizing a hybrid framework that leverages the strengths of two advanced models: a Feature Tokenizer Transformer (FT-Transformer) and a Gradient Boosted Decision Tree (GBDT). The FT-Transformer was selected for its superior ability to capture intricate feature interactions and contextual relationships within the high-dimensional data through its self-attention mechanisms. The GBDT model complements this by providing exceptional predictive power on tabular data and robustness against missing values. Both models achieved a notable accuracy of 93%, demonstrating their individual efficacy. This study concludes that the application of these models on a large scale offers a highly accurate and reliable tool for clinicians, potentially significantly improving preventive cardiology strategies and patient outcomes.
"Heart Disease Prediction Using Machine Learning and Data Mining Techniques ", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.10, Issue 9, page no.b237-b245, September-2025, Available :https://ijsdr.org/papers/IJSDR2509131.pdf
Volume 10
Issue 9,
September-2025
Pages : b237-b245
Paper Reg. ID: IJSDR_305019
Published Paper Id: IJSDR2509131
Downloads: 00060
Research Area: Science All
Country: Mumbai, Maharashtra, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
