Every employee in the company that deals with data needs to have clean, noise-free data. Since data warehouses store and update enormous amounts of data from several sources, there is a potential that some of those references may contain inaccurate data. Due to the noise, inefficacy, and poor characterization of the vast amount of accessible data, as well as the ensuing insensitivity and inefficiencies of human data cleaning and labeling, the presentation of the data has become ambiguous, and the assessment of the information has become difficult. A hole in the creation of a better data analysis method was identified. This helped to guide the creation of a Python script for automatically cleaning and labeling data. The first step in the strategy used in this study to accomplish its goals and objectives was to obtain a financial dataset from the top database, "Kaggle". Create a machine learning (ML) approach in Python that intends to automate the financial dataset cleaning. This covers ingesting data, addressing incomplete data, addressing anomalies, one-hot wrapping and label encoding, extracting date and time values, and data normalization. Implementing an unsupervised machine learning method that attempts to automate financial dataset labeling (k-means). Using the method includes the elbow principle, k-means clustering, data modeling of "age" versus "arrival," dimensionality reductions, computer vision, and dataset categorizing using the groupings. Empirical assessment of the cleaned and labeled automated trading dataset utilizing a comparison of the cleaned dataset before and after PCA adoption. The results show that the developed ML technique not only improved the performance of the audit data used in this study, but it also classified the data after cleaning it and removing the unpleasant section and incomplete data, as shown by the k-means segmentation result and grouping by PCA
26 June 2025 in International Journal of Medical Informatics
Acute Lymphoblastic Leukemia (ALL) is the most common pediatric cancer, requiring early detection to save lives and reduce the financial burden of advanced-stage treatment. While traditional diagnostic methods are time-consuming and resource-intensive, deep transfer learning offers a computationally efficient alternative for medical image classification. This study employed two widely recognized transfer learning algorithms, VGG-19 and EfficientNet-B3, to detect ALL using a publicly available dataset of 10,661 images from 118 patients. Data preprocessing included resizing, augmentation, and normalization. The models were trained for 100 epochs, with batch sizes of 30 for VGG-19 and 32 for EfficientNet-B3. Evaluation metrics such as accuracy, precision, recall, and F1 score were used to assess model performance. Statistical significance testing was performed using paired t-tests (p < 0.05). Comparative analysis was performed with existing studies to validate the findings. EfficientNet-B3 significantly outperformed VGG-19, achieving an average accuracy of 96 % compared to 80 % for VGG-19 (p < 0.001). EfficientNet-B3 demonstrated superior performance in handling class imbalance, with the minority class (Hem) achieving precision, recall, and F1 scores of 97 %, 89 %, and 93 %, respectively. VGG-19 struggled with the minority class, achieving lower recall (51 %) and F1 score (62 %). However, dataset limitations including single-source origin may affect generalizability. This study highlights the effectiveness of EfficientNet-B3 as a reliable tool for early ALL detection, offering high accuracy and computational efficiency. Clinical implementation requires addressing computational constraints and integration challenges. Future research could integrate multimodal datasets to identify risk factors and further improve diagnostic accuracy.
The application of Artificial Intelligence (AI) to diverse field has been widely accepted ranging from transportation, education, logistics, entertainment and health. Specifically, in recent time, the application of Machine Learning (ML) a subset of AI has equally got wide acceptance and relevance in various aspect of medicine ranging from diagnosis and prediction of diseases, development of drugs and treatment plan among others. This has made medical procedure to be faster, more accurate and easier than using the traditional approach. Stroke is reported to be one of the major causes of death and this can be reduced by studying the risk factors causing it and predicting its occurrence so as to educate people about it. This study developed a stroke prediction system with a modified Gated Recurrent Unit (GRU), a structured stroke dataset was gotten from Kaggle, which went through different preprocessing techniques like label Encoder, Normalization with MinMax, dropping of irrelevant values. Furthermore, different data balancing techniques were employed to improve the accurate performance of the model. The preprocessed dataset was used by GRU for prediction. The system gave average accuracy, Area Under Curve (AUC) and prediction time of 80.42%, 0.8940 and 0.678 s respectively. The developed system outperformed other ML algorithms like LSTM, GRU-LSTM, Support Vector Machine (SVM) and Logistic Regression. This study showed that GRU a variant of Recurrent Neural Network, could give better predictive performance on structured data rather than only with streaming data; therefore, showcasing the improved performance of GRU over other variants of RNN.
Gas leakage is a major concern to commercial, industrial, residential locations and gas-powered transportation vehicles in Nigeria and in some other developing nations of the world. The havoc wrecked by gas leakage has not ceased to be common news on media. Therefore, it is essential to device measures of intelligently detecting and controlling gas leakage, hence reducing the possibility of fire outbreak in homes and industries. This study presents portable, effective and cost-efficient preventive measure to avoiding danger associated with gas leakage and fire outbreak in homes and industries through the design and construction of smart gas leakage detector with audio-visual alarm and automatic shut off control.MQ 3 gas sensor detects gas leakage, and the microcontroller triggers four output means for alerts and one control measure , that is through a buzzer which gives audible sound and a Light Emitting Diode (LED) which turns ON (red visible light); the LCD status display changes from the default "NO GAS IS DETECTED" to "GAS HAS BEEN DETECTED"; also, the GSM module dials the pre-registered mobile number to alert an off premises emergency service, while the electric valve automatically switches off the gas main source as a control measure to prevent further gas leakage. With this design, gas leakage can be intelligently monitored and controlled while hazards relating to gas leakage are ultimately prevented. However, when no gas leakage is detected, the buzzer and the light emitting diode remain off, hence it does not create nuisance to the environment. To this end, this project has solved a major problem in our nation Nigeria by presenting a portable, efficient, cost effective, smart GSM based gas detector to aid human ability to monitor, control and secure human lives and properties.