Journal of Data Mining and Management

Enhancing Accident Risk Prediction with Novel Data and Findings from Heterogeneous Sparse Sources

S M Mustaquim — 2024-02-05

Accurate accident risk prediction is paramount for proactive safety measures and resource allocation. This paper introduces an innovative approach to enhance accident risk prediction by leveraging novel data sources and uncovering insights from heterogeneous sparse data. Traditional models often suffer from limitations in data diversity and scope, hindering their predictive capabilities. In response, our study integrates a wide range of heterogeneous data, including traffic flow data, weather conditions, road infrastructure, and historical accident records. To address the challenges of working with sparse data, we employ advanced data science techniques, including feature engineering, imputation, and machine learning. The paper presents a new dataset that combines diverse data types, providing a comprehensive foundation for our predictive model. Through rigorous analysis, we extract valuable insights from these heterogeneous sources to improve accident risk assessment. The proposed approach offers several advantages, including the ability to predict accidents in previously underrepresented areas and under varying conditions. We evaluate the model's performance through extensive experimentation and validate its accuracy against real-world accident data. Our findings demonstrate significant enhancements in prediction accuracy compared to conventional models. This research contributes to the field of accident risk prediction by showcasing the potential of heterogeneous sparse data integration and advanced data science techniques. It highlights the importance of utilizing novel data sources and the value of uncovering hidden patterns and insights to foster safer environments and more efficient resource allocation in accident-prone areas.

A Survey on Early Cattle Diagnostic Techniques

Mohan Raju N — 2024-02-16

Dairy cattle's well-being is the rancher's riches. One of the huge dangers to dairy ranchers is cow infections. Assume the illnesses are not distinguished at the right time some other time when the infection turns into a persistent stage. The treatment cost will be more, and further, the monetary expense caused to the rancher is irreversible as the dairy cattle may not give the yield it was offering before, even after getting mended. The thought is to recognize the steers’ illnesses in the beginning phase so the rancher can take the treatment measures immediately. The study is centred on applying machine learning to automatically detect haemoprotozoan illness in cattle. By combining cutting-edge technology in data analysis, the project seeks to produce a reliable system that uses sophisticated algorithms for accurate and effective diagnosis. Improvements in early detection capabilities lead to better management of cattle health and increased output in agriculture.

Develop Therapy Materials in Hindi for Misarticulation Children

Ayush Deshmukh — 2024-03-22

The Hindi Articulation Therapy App is an invaluable tool for addressing speech recognition challenges in India. Its intuitive interface caters to professionals, children, and parents alike, making it accessible to all. The "Level Project" feature offers fundamental instructions on mouth and tongue positioning, akin to the initial step in speech therapy. With the "View Library," users access a diverse array of resources, enriching Hindi-speaking practice through visual aids and authentic audio recordings. For effective therapy sessions, the "Articulation Practice" feature provides ample speaking exercises. Parents stay informed about their child's progress through timely notifications, fostering collaboration in the treatment process. The "Pronunciation Videos for Kids" aid children in mastering sounds, while the "Pronunciation Practice" feature offers multiple methods for refining pronunciation skills. Ultimately, this app serves as a comprehensive tool for enhancing Hindi language proficiency and communication skills. Its versatility and user-centric design make it an indispensable asset for individuals seeking to improve their Hindi articulation and communication abilities.

A Machine Learning Approach for Predicting Plant Diseases and Ensuring Crop Health

Rahul A Patil — 2024-04-10

The widespread prevalence of plant diseases poses significant challenges for the agricultural industry in maintaining crop health and ensuring the continued sustainability of food production. Early detection and accurate diagnosis of such diseases are critical for timely and effective intervention and treatment. Over the last few years, machine learning techniques, specifically Convolutional Neural Networks, have generated promising results in various applications, in particular, computer vision tasks. This paper proposes an innovative method, termed DeepPlantDx, which utilizes CNN algorithms to accurately predict plant diseases and, eventually, afford sustainable crop management strategies. In this paper, we demonstrate effective plant disease diagnosis using convolutional neural networks-based pre-trained models. We focus on hyperparameter tunning of popular pre-trained models, namely DenseNet-121, ResNet-50, VGG-16, and Inception V4. Experiments were performed with the benchmark dataset Plant Village, which contains 54,305 unique image datasets with 38 categories of different plant diseases. Furthermore, the paper analyzes DeepPlantDx's possible impacts such as boosting crop yields, cutting down on the use of pesticides, and promoting sustainable agriculture techniques.

Detecting Anomalies in IOT Devices through Machine Learning Techniques

Dattatray G. Takale — 2024-04-30

The rapid increase in Internet of Things devices has sparked fears of vulnerability in various areas, including anomaly detection. This paper introduces a novel use of machine learning algorithms, Support Vector Machines and Random Forests, as well as ensemble methods such as stacking and voting classifiers for anomaly detection on the Internet of Things. Based on the NSL-KDD dataset, the experiment demonstrates the efficacy of RF and stacking classification methods, achieving high accuracy with fewer false positives than current literature. Both ensemble methods, Voting Classifier RF + AB and Stacking Classifier RF + MLP with LightGBM, achieve exceptional performance, recall and precision, proving suitable for identifying and managing anomalous behaviours in various systems. Moreover, the project includes integrating user evaluation through a Flask framework front-end with user authentication, a critical component of IoT anomaly detection's practical implementation. This paper demonstrates the potential of ML approaches in improving the security and endurance of the Internet of Things by efficiently identifying and managing variations in multitudes of Internet of Things working environments.