Smartic: A smart tool for Big Data analytics and IoT

• DOI: 10.12688/f1000research.73613.2
• Publication Date: 2024-02-06T12:25:07Z
• Authors: Shohel Sayeed; Abu Fuad Ahmad; Tan Choo Peng
• Keywords: Open peer review

<ns3:p>The Internet of Things (IoT) is leading the physical and digital world technology to converge. Real-time massive scale connections produce a large amount versatile data, where Big Data comes into picture. refers large, diverse sets information with dimensions that go beyond capabilities widely used database management systems, or standard data processing software tools manage within given limit. Almost every big dataset dirty may contain missing mistyping, inaccuracies, many more issues impact analytics performances. One biggest challenges in discover repair data; failure do this can lead inaccurate results unpredictable conclusions. Different imputation methods were employed experimentation various value techniques, performances machine learning (ML) models compared. A hybrid model integrates ML sample-based statistical techniques for being proposed. Furthermore, continuation involved best imputation, chosen based on performance subsequent feature engineering hyperparameter tuning. K-means clustering principal component analysis applied our study. Accuracy, evaluated outcome, improved dramatically proved XGBoost gives very high accuracy at around 0.125 root mean squared logarithmic error (RMSLE). To overcome overfitting, K-fold cross-validation was implemented.</ns3:p>