Volume 13 | Issue 5 |

 Abstract

Ensuring the quality of tabular data is a critical preprocessing step before training reliable predictive models. We present a web-based application built with Django that automates data quality assessment and cleaning on uploaded CSV datasets, followed by training of a supervised learning model. The system ingests user-provided CSV files and analyzes issues such as missing values, outliers, and skewed features, providing visual dashboards summarizing key data quality metrics. Automated cleaning operations (mean/mode imputation, outlier clipping via the IQR rule, log-transform for skewed features, and one-hot encoding of categoricals) are performed using Python libraries (Pandas, SciPy, scikit-learn), producing a cleaned dataset ready for analysis. Users can then select a target variable and train multiple regression/classification algorithms (e.g. linear regression, random forest, k-NN), with the best model highlighted by evaluation metrics (e.g. mean squared error). In a case study on a housing price dataset, the application identified approximately 5.97% missing values, 1.47% outliers, and 25.0% highly skewed features, yielding a composite data quality score of 67.56%. A Random Forest regressor achieved the lowest error among models, consistent with known performance. The interface displays feature correlations and per-unit effect sizes, along with interactive charts. This integrated solution streamlines the machine learning pipeline by combining statistical data cleaning and model training in a single Django-based framework.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Data Quality, Data Cleaning, Django, Machine Learning, Predictive Model, CSV Data, Web Application.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This study investigated the impact of security on the academic achievement of school-going students in Haryana, India. The objective was to compare academic performance across varying levels of security and examine the relationship between security and academic success. The hypotheses proposed that there would be no significant differences or relationships between these variables. A descriptive survey design was used, focusing on 10th-grade students from four districts in Haryana. A sample of 800 students was selected through stratified random sampling. The Shah Security Insecurity Scale (2005) assessed perceived security, while academic achievement was measured using students' previous academic scores. Data were analyzed using descriptive and inferential statistics, including mean comparison and correlation analysis. The findings revealed that students with a higher sense of security demonstrated better academic performance, and a strong positive relationship existed between security and academic success. These results underline the critical role of a secure environment in fostering students' academic potential. The study recommends implementing robust security measures in schools, offering psychological support, and training teachers and parents to recognize and address insecurity.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Security, Insecurity, Academic Achievement, School-Going Students, Education Policy

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents a real-time ECG-based heartbeat monitoring application designed to predict a patient's health condition remotely without the need for constant medical supervision. The system uses an ECG sensor to acquire the patient's heartbeat signals, which are then transmitted to a cloud database. A machine learning model processes this data to predict possible abnormalities and provides feedback through a mobile or web interface. This application enables users and caretakers to monitor heart health anytime, anywhere, improving early detection and reducing dependency on immediate clinical consultation.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

ECG, IoT, Heartbeat Monitoring, Health Prediction, Machine Learning, Remote Health Monitoring.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Background and Objectives: Cryptographic hash functions are the linchpins of mobile services, blockchains, and many other technologies. Designing cryptographic hash functions has been approached by research communities from the physics, mathematics, computer science, and electrical engineering fields. The emergence of new hash functions, new hash constructions, and new requirements for application-specific hash functions, such as the ones of mobile services, have encouraged us to make a comparison of different hash functions and propose a new classification. Methods: Over 100 papers were surveyed and reviewed in detail. The research conducted in this paper has included four sections; article selection, detailed review of selected articles, data collection, and evaluation of results. Data were collected as new hash function properties, new hash function constructions, new hash function categories, and existing hash function attacks which are used to evaluate the results. Results: This paper surveys seven categories of hash functions including block cipher-based functions, algebraic-based functions, custom-designed functions, Memory-hard Functions (MHFs), Physical Unclonable Functions (PUFs), quantum hash functions and optical hash functions. To the best of our knowledge, the last four mentioned categories have not been sufficiently addressed in most existing surveys. Furthermore, this paper overviews hash-related adversaries and six hash construction variants. In addition, we employed the mentioned adversaries as evaluation criteria to illustrate how different categories of hash functions withstand the mentioned adversaries. Finally, the surveyed hash function categories were evaluated against mobile service requirements. Conclusion: In addition to new classification, our findings suggest using PUFs with polynomial-time error correction or possibly bitwise equivalents of algebraic structures that belongs to post-quantum cryptography as candidates to assist mobile service interaction requirements.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Optical hash function Memory-hard function Bandwidth-hard function Physical unclonable function Quantum hash function Application-specific hash function

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Gastrointestinal (GI) Hemorrhage is a critical medical condition marked by bleeding within the digestive system, which can occur at any point along the alimentary tract. It can be caused by factors such as peptic ulcer disease, inflammatory bowel disease (IBD), or excessive use of nonsteroidal anti-inflammatory drugs (NSAIDs). Early detection is crucial to prevent morbidity and mortality, but existing diagnostic modalities, including endoscopy and imaging studies, are invasive, costly, and uncomfortable for patients. This project proposes a non-invasive diagnostic system utilizing bioimpedance spectroscopy to detect GI hemorrhage. The system involves placing surface electrodes on the abdominal wall, where a small alternating current (AC) signal is used to measure tissue impedance. The presence of blood in the GI tract alters the impedance characteristics, which are detected by the impedance sensor and processed in real-time using a microcontroller-based unit, such as Arduino. he designs ensures rapid, painless, and accurate detection without the need for hospitalization. The proposed system aims to improve patient outcomes by offering a user-friendly, portable, and cost-effective solution for early diagnosis. It is especially beneficial in home-care settings and for patients who require frequent monitoring. Future developments include integrating artificial intelligence (AI) to enhance diagnostic accuracy and expanding the system's application to detect other gastrointestinal disorders non-invasively. SMART Medical Diagnostic is changing the way we detect gastrointestinal bleeding by using a simple, non-invasive method. With impedance sensor technology, it quickly and safely finds internal bleeding without the need for uncomfortable procedures. This new approach gives doctors accurate results in real-time, making diagnosis faster and easier for patients. SMART Medical Diagnostic is the future of gastrointestinal care--smart, simple, and patient-friendly. Keywords: Gastrointestinal (GI) hemorrhage, Gastrointestinal bleeding, Alimentary canal, Peptic ulcer disease, Inflammatory bowel disease (IBD), Nonsteroidal anti-inflammatory drugs (NSAIDs), Early detection, Morbidity and mortality, Non-invasive diagnostic system, Bioimpedance spectroscopy, Tissue impedance, Surface electrodes, Abdominal wall sensing

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords: Gastrointestinal (GI) hemorrhage, Gastrointestinal bleeding, Alimentary canal, Peptic ulcer disease, Inflammatory bowel disease (IBD), Nonsteroidal anti-inflammatory drugs (NSAIDs), Early detection, Morbidity and mortality, Non-invasive diagnostic system, Bioimpedance spectroscopy, Tissue impedance, Surface electrodes, Abdominal wall sensing

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Vehicle overloading presents critical challenges to infrastructure maintenance and traffic safety. Current static weighing systems suffer from operational inefficiencies, prompting the development of an intelligent real-time monitoring solution. This research introduces an IoT-based overload detection system utilizing precision load cells and edge processing for instantaneous weight analysis. The system addresses three key consequences of overloading: compromised vehicle stability, accelerated mechanical wear (chassis, suspension, steering), and regulatory non-compliance. Implementation results demonstrate 92.3% detection accuracy with automated alerts via integrated displays and remote notifications. The solution significantly improves upon traditional methods by enabling proactive load management during cargo operations, simultaneously enhancing safety, compliance, and operational efficiency in commercial transport.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

vehicle overloading, real-time monitoring, IoT system, load cells, edge processing, weight detection, traffic safety, infrastructure protection, regulatory compliance, intelligent transportation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This study is about the use of Variable Frequency Drives in Direct Current to Alternating Current applications. A Variable Frequency Drive (VFD) primarily manages the speed and torque of an AC motor by modifying the frequency and voltage it receives. However, their role in DC-to-AC conversion presents new concepts of power control for improvement in system efficiency and adaptability in systems that demand versatility within electrical outputs. A Variable Frequency Drive (VFD) is a key component in modern electrical systems, used to regulate the speed and torque of motors by altering the input frequency and voltage.This study focuses on the design and selection of VFDs, particularly for industrial applications, with emphasis on the key components such as diodes in the rectifier stage and IGBTs in the inverter stage. The paper also includes the detailed calculation for selecting VFD components based on motor specifications, using a 5 HP (3.73 kW) motor as a case study. The main objective is to offer insights into the appropriate VFD ratings, component selection, and the application of VFDs for controlling motors efficiently.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project aims to develop a virtual try-on system for shirts using computer vision techniques in Python. The system will leverage the OpenCV library to capture live video streams from a webcam and apply body gesture recognition to track key points on the user's body. By overlaying virtual shirts onto the user's video feed in real-time, the system will simulate a realistic try-on experience. The project will focus on enhancing the user's online shopping experience by providing a visual representation of how different shirts look and fit. The system's interface will allow users to interactively select and try on various shirts, providing them with a personalized and engaging shopping experience. Evaluation of the system will be based on its accuracy in simulating shirt fit, user feedback on usability and realism, and its ability to seamlessly integrate with existing e-commerce platforms. The outcome of this project is expected to improve user satisfaction and increase confidence in purchasing shirts online. Additionally, we identify and discuss similarities and differences in these methods.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

This project focuses on developing a real-time virtual try-on system for shirts using Python and OpenCV, incorporating body gesture recognition and keypoint tracking. Keywords include virtual try-on, computer vision, OpenCV, real-time overlay, body tracking, e-commerce integration, and user experience enhancement.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract Background: Hamstring injuries are common among athletes and can be particularly challenging to manage in para-athletes with underlying medical conditions. Thalassaemia minor may influence tissue healing, necessitating specialized approaches in rehabilitation. Aim: To evaluate the effectiveness of Growth Factor Concentration (GFC) therapy as a complementary intervention in the physiotherapeutic management of a hamstring injury in a T47 para-athlete with thalassaemia minor. Methodology: This case study involved a 31-year-old female T47 para-athlete diagnosed with a semitendinosus tear confirmed via ultrasound. Initial conservative treatment showed limited improvement. GFC therapy was introduced alongside a structured physiotherapy regimen. Functional and clinical assessments were conducted before and after treatment, including range of motion tests, manual muscle testing, pain scales, and balance evaluation. Results: Post-treatment results demonstrated significant improvement: hamstring strength improved from Grade 3/5 to 5/5, pain reduced from 8/10 to 1/10, and flexibility and stability were fully restored. The athlete returned to sport-specific activity without recurrence of symptoms. Conclusion therapy, when combined with targeted physiotherapy, significantly enhanced the healing process and functional recovery in a para-athlete with complex clinical considerations. It may be considered a viable adjunct in sports rehabilitation protocols, particularly for cases with compromised healing potential.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords: Hamstring strain, para-athlete, Growth Factor Concentration, thalassaemia minor, physiotherapy, soft tissue injury, sports rehabilitation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Autoimmune joint diseases are characterized by chronic synovial inflammation and immune dysfunction, leading to joint destruction and impaired function. Integrative approaches such as Ayurveda may offer therapeutic potential through multi-targeted strategies addressing symptoms and immune modulation. This study evaluated the clinical effectiveness and immunomodulatory impact of the Researchayu Regimen for Autoimmune Treatment (RRAT) in patients with seropositive autoimmune joint diseases. An open-label, retrospective, single-arm, multicenter study was conducted at five ResearchAYU centers across Maharashtra, India. A total of 55 patients aged 40-85 years with knee pain and positive rheumatoid factor (RA factor) and/or antinuclear antibody (ANA) were included. The RRAT protocol involved structured Ayurvedic oral formulations, topical Shuddhu Lepa, and therapeutic purgation (Sadhyovirechana). Clinical assessments and laboratory investigations (RA factor, ANA, C-reactive protein [CRP], and serum uric acid) were performed at baseline and after 90 days. Statistical analysis included descriptive statistics, McNemar's test for seroconversion, and paired t-tests or Wilcoxon signed-rank tests for continuous variables, with significance at P < 0.05. At day 90, 80.64% of ANA-positive and 81.25% of RA factor-positive participants achieved seroconversion. CRP and serum uric acid levels showed significant reductions of 33.23% and 25.05%, respectively. Clinically, there was a 61.22% mean relief in walking-related pain and a 57.5% resolution of morning stiffness. No serious adverse events were reported. These findings suggest that the RRAT protocol may offer beneficial clinical and immunological effects in autoimmune joint diseases. Prospective randomized controlled trials are needed for further validation.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Autoimmune joint disease, Rheumatoid arthritis, Ayurveda, Researchayu Regimen for Autoimmune Treatment (RRAT), Immunomodulation, C-reactive protein, Seroconversion.

  License

Creative Commons Attribution 4.0 and The Open Definition