IMPROVING SLEEP DISORDER DIAGNOSIS THROUGH
OPTIMIZED MACHINE LEARNING APPROACHES
ABSTRACT
Classifying sleep disorders, such as obstructive sleep apnea and insomnia, is
crucial for improving human quality of life due to their significant impact on
health. The traditional expert-based classification of sleep stages, particularly
through visual inspection, is challenging and prone to errors. This fact highlights
the need for accurate machine learning algorithms (MLAs) for analyzing,
monitoring, and diagnosing sleep disorders. This paper compares the MLAs for
sleep disorder classification, specifically targeting None, Sleep Apnea, and
Insomnia, using the Sleep Health and Lifestyle Dataset. We conducted two
experiments. In the first one, we selected five key features from the feature
spaces using the Gradient Boosting Regressor based on the Mean Decrease
Impurity (MDI) technique. We chose two key features using the same
methodology in the second experiment. We utilized 15 machine learning
classifiers, and Gradient Boosting, Voting, Catboost, and Stacking Classifiers
achieved an identical classification accuracy of 97.33%, with Precision, Recall, F1-
score of 0.9733, and Specificity of 0.9569 in the original feature space. Among
these, Gradient Boosting had the highest AUC of 0.9953 and was 3.36, 5.86, and
20.16 times faster than Voting, Catboost, and Stacking Classifiers, respectively.
In the second experiment, the Decision Tree achieved the highest accuracy of
96% in the original and engineered feature spaces and was 149.33 times faster
in the engineered feature space. Thus, this research proposes Gradient
Boosting as the most effective method, outperforming all state-of-the-art
techniques by achieving the highest accuracy, precision, recall, F1-score, and
AUC, highlighting its superior classification performance and computational
efficiency.
TEAM MEMBERS:
1. J. Vasanth Kumar (Y22CY3224)
2. K. Shanmukha varma (Y22CY3229)
3. K. Suresh (Y22CY3232) UNDER THE GUIDANCE OF:
4. B. Tejendra Sai (Y22CY3208) Dr. Balaji Vicharapu
B. Tech., M. Tech., Ph. D., M.B.A.
ENHANCING DEFECT CLASSIFICATION IN SOLAR PANELS
WITH ELECTROLUMINESCENCE IMAGING AND ADVANCED
MACHINE LEARNING STRATEGIES
ABSTRACT
Electroluminescence (EL) imaging is the most widely used diagnostic technique
for identifying flaws at every stage of the production, installation, and operation
of solar modules. This method can potentially reduce power outages by
locating and fixing solar module faults such microcracks and breaks in the
finger lines. The EL test is a reliable inspection method, however, because of
complex fault patterns and heterogeneous backgrounds, interpreting EL
images can be difficult. As a result, assessing damaged cells and determining
the severity of an issue necessitates specialized knowledge, which makes
manually executing these methods for each cell time-consuming. Because of
this, automated visual inspection of solar cells becomes very important. In this
work, a novel system for automatically identifying and categorizing solar cell
faults is presented. A strong CNN model created from scratch is used to extract
deep features. Utilizing the recently developed RSWS classification method, the
deep characteristics are evaluated. The popular ELPV dataset with two and
four classes is used to test the suggested methodology. For the two-class
classification problem, the classification performance is 98.17%, and for the four-
class classification problem, it is 97.02%.
TEAM MEMBERS:
1. J. Vasanth Kumar (Y22CY3224)
2. K. Shanmukha varma (Y22CY3229)
3. K. Suresh (Y22CY3232) UNDER THE GUIDANCE OF:
4. B. Tejendra Sai (Y22CY3208) Dr. Balaji Vicharapu
B. Tech., M. Tech., Ph. D., M.B.A.
PREDICTING THE CLASSIFICATION OF HEART FAILURE
PATIENTS USING OPTIMIZED MACHINE LEARNING
ALGORITHMS
ABSTRACT
Heart failure is a critical condition with a high mortality rate, making accurate
survival prediction essential for timely interventions. This study proposes an
optimized machine learning approach using Gradient Boosting Machine (GBM)
and Adaptive Inertia Weight Particle Swarm Optimization (AIW-PSO) to predict
heart failure survival. The dataset, sourced from Kaggle, includes clinical
features such as age, ejection fraction, and serum creatinine levels for 299
heart failure patients. To address the imbalance in survival outcomes, Synthetic
Minority Over-sampling Technique (SMOTE) was employed to balance the
dataset, followed by SelectKBest and Chi-square feature selection methods to
retain the most significant predictors. The optimized hyperparameters for the
GBM model were identified using the AIW-PSO algorithm, which effectively
balanced exploration and exploitation by adaptively adjusting inertia weights.
Model selection was further refined using information criteria, including Akaike
Information Criterion (AIC) and Bayesian Information Criterion (BIC), ensuring
that the best-performing model was chosen based on both predictive
accuracy and model complexity. The optimized GBM model achieved a test
accuracy of 94%, demonstrating superior performance compared to traditional
machine learning models. The study underscores the importance of
hyperparameter tuning through metaheuristic algorithms and highlights the
potential of AIW-PSO in enhancing model performance for clinical prediction
tasks. These findings have significant implications for clinical decision-making,
offering a reliable and interpretable tool for predicting patient outcomes in
heart failure management.
TEAM MEMBERS:
1. J. Vasanth Kumar (Y22CY3224)
2. K. Shanmukha varma (Y22CY3229)
3. K. Suresh (Y22CY3232) UNDER THE GUIDANCE OF:
4. B. Tejendra Sai (Y22CY3208) Dr. Balaji Vicharapu
B. Tech., M. Tech., Ph. D., M.B.A.
ENHANCING PHISHING DETECTION: A MACHINE
LEARNING APPROACH WITH FEATURE SELECTION AND
DEEP LEARNING MODELS
ABSTRACT
With the rise in cybercrime, phishing remains a significant concern as it targets
individuals with fake websites, causing victims to disclose their private
information. The effective implementation of phishing detection relies on cost
efficiency, with the increased feature extraction factor contributing to these
costs. This research analyzes a dataset containing 58,645 URLs, examining 111
features of the latest phishing websites dataset to identify the differences
between phishing sites and legitimate sites. Astonishingly, using only 14
characteristics, the feedforward model achieved a remarkable accuracy of
94.46%, confirming the efficiency of Machine Learning in phishing detection.
Through the exploitation of a multiple classifier collection, including Deep Neural
Network (DNN), Wide and Deep, and TabNet, this research advances ongoing
efforts to improve the accuracy and efficiency of phishing detection
mechanisms and enhance cybersecurity defenses against malicious activities.
The methodology introduces a new metric called the ‘anti-phishing score,’
which evaluates performance based on false positives and negatives, beyond
traditional model accuracy. The model was trained through a robust design of
extensive experimentation and hyperparameter-sensitive grid search, ensuring
an optimized configuration for phishing detection. Furthermore, the trained
model was validated on a new dataset to evaluate its generalizability,
enhancing its practical applicability.
TEAM MEMBERS:
1. J. Vasanth Kumar (Y22CY3224)
2. K. Shanmukha varma (Y22CY3229)
3. K. Suresh (Y22CY3232) UNDER THE GUIDANCE OF:
4. B. Tejendra Sai (Y22CY3208) Dr. Balaji Vicharapu
B. Tech., M. Tech., Ph. D., M.B.A.