International Journal of Advanced Research in Artificial Intelligence and Machine Learning Reviews

From Crisis to Care... AI Revolutionising Humanitarian Healthcare

Udaiveer Singh Brar — 2026-01-22

Background: Humanitarian healthcare and disaster response greatly rely on quick, informed and coordinated actions to save lives and resources.Artificial intelligence (AI) is coming out as a powerful tool that has the capability to support crisis management through predictive analytics, decision support, and efficient allocation of medical resources.
Objectives: To explore the transformation of artificial intelligence in disaster response, emergency care and equitable healthcare systems.
Methods: The literature was reviewed for recent developments in the domain of AI applications for disaster preparedness and emergency healthcare, identifying persistent challenges such as limited datasets, ethical dilemmas and integration issues.
Results: This study highlighted that AI usage in humanitarian healthcare improves disaster response efficiency and enhances predictive accuracy. However, real-time system integration highlights the need for more transparent and collaborative AI frameworks.
Conclusion: AI has immense potential for transforming humanitarian healthcare with faster, data-driven, and equitable crisis response. Focus on ethical governance is required to realise its impact in real-life disaster scenarios.

Evaluating Machine Learning Algorithms for Automated Personality Judgment

Rajiv Kumar — 2026-01-22

Traditional methods of personality judgement, such as self-report questionnaires and manual assessments, are often limited by subjectivity, time consumption and vulnerability to social desirability bias. These drawbacks highlight the need for automated and data-driven techniques that can provide more objective and scalable personality evaluation. In this study, we explore the use of machine learning (ML) algorithms to predict personality traits and systematically compare their performances. Models including linear regression, decision tree, random forest, support vector machine (SVM) and AdaBoost are implemented on a benchmark dataset. The algorithms are evaluated using standard metrics such as accuracy, precision, recall and F1-score to ensure a comprehensive analysis. Results reveal distinct strengths and weaknesses across classifiers, offering insights into the most effective approaches for personality judgement. The findings demonstrate the potential of ML in advancing personality assessment and provide a foundation for building reliable, interpretable and scalable solutions. Such approaches can be applied in domains like human resource management, education and mental health, where accurate personality insights are essential for informed decision-making and personalised interventions.

Smart Hydroponic System Using IoT and Machine Learning for Water Quality Monitoring

Priyanka Kumari — 2026-01-17

Hydroponics has gained prominence as a sustainable and resource-efficient alternative to traditional soil-based agricul ture. However, maintaining optimal water quality in hydroponic systems requires constant monitoring of critical parameters such as pH, electrical conductivity (EC), and temperature. This paper presents a smart hydroponic monitoring system that integrates IoT-enabled sensors with machine learning algorithms to provide real-time data acquisition and water quality classification. The proposed hydroponic system, built on an Ebb and Flow configu ration, uses a Raspberry Pi 5 as the central controller to process data from pH, EC, and temperature sensors. A comprehensive dataset was collected and labeled based on established water quality thresholds to train and evaluate several classification models, including Logistic Regression, Random Forest, Support Vector Machine (SVM), and XGBoost. The XGBoost model achieved perfect performance (100% accuracy, precision, recall, and F1-score) in classifying water conditions as â€™Safeâ€™ or â€™Unsafeâ€™. The resulting model was serialized for deployment, enabling real time inference on the edge device. The work demonstrates a scalable and cost-effective framework for enhancing automation in hydroponic farming, aiming to improve plant health, optimize nutrient management, and minimize human intervention through intelligent, data-driven decision-making.

AI-Driven Precision Farming: A Scalable Neural Approach for Crop Prediction

Bikramaditya Chakraborty — 2026-01-22

Precision farming is an application that uses infor mation to maximize crop production and resource use. The paper states that we have JK soil analysis data and it has been trained on a machine learning model to give the prediction of the crops. Environmental factors (Rainfall, Temperature) were evaluated and soil properties (pH, N, P, K, Zn, Fe, Mn). In order to achieve a high data quality and reliability, sophisticated preprocessing methods were used. Z-score-based filtering was used to eliminate outliers and the Min-Max scaling was used to standardize the input space by normalizing numerical features. The Chi-Squared test was used to reduce the most significant 10 features to further do the prediction. The preprocessing pipeline was thorough and the noise and redundancy were reduced to provide a solid dataset to train the model. The neural network was created using two hidden layers (64 and 32 neurons) and attained 98.45 percent test accuracy. Other important contributions are strong outlier management, environmental data incorporation and scalability in newer datasets. The method provides a basis of real-time and scalable accurate farming frameworks, and possible utilizations of pest and fertilizer management.

Behavioral Segmentation of Black Friday Consumers

Nirmaljeet Singh — 2026-01-22

This study explores Black Friday sales by looking at customer behaviour, store dynamics, and sales patterns, with a focus on how these factors affect international retail events both offline and online. The study reveals customer preferences impacted by marketing tactics, product availability, and discounts using statistical analysis, predictive modelling, and historical sales data. By utilising insights into consumer behaviour during high-demand times like Black Friday, the findings seek to help businesses improve marketing strategies, improve customer experiences, and increase revenue. In the end, they provide useful ideas for optimising sales performance.

Exploring the Role of Programming and Cognitive Skills in Code Comprehension and Workload Measurement

Divjot Singh — 2026-01-22

This study explores how programming and cognitive skills contribute to software com- prehension and how programmers’ cognitive workload is measured. The review shows that code reading, tracing, and debugging are the most frequently studied programming skills, supported by cognitive abilities such as working memory, reasoning, and problem-solving. Coding tasks and comprehension tests are the most common evaluation methods, while advanced tools such as fMRI, EEG, and eye-tracking provide deeper insights into mental effort. Key parameters used across studies include task accuracy, completion time, and brain activity. However, research in this area still faces challenges such as small sample sizes, self-report bias, high sensor costs, and difficulty replicating real programming condi- tions. Overall, the findings highlight the need for practical and scalable methods to better understand how programmers think and manage cognitive load while working with complex code.

Forecasting Air Quality Index (AQI) Using Machine Learning Models: A Comparative Study

Prabhkiran Kaur — 2026-01-22

One of the biggest environmental issues affecting our health and well being is air quality, an unseen danger that we breathe every day. Packed with dangerous gases and microscopic particles, poor air quality is a silent killer that can cause anything from chronic respiratory issues to serious, life-threatening diseases. The enormity of this issue emphasises how urgently we need information that is easy to understand in order to safeguard our communities. By concentrating on the Air Quality Index (AQI), a simple method of expressing how clean or polluted the air is, this research directly addresses that challenge. The goal is to uncover the hidden narrative within the numbers by employing intelligent computer models (machine learning) to sort through years’ worth of air pollution data, includingdaily readings of smog, soot, and other pollutants. The objective is to increase the usefulness of the AQI by creating a system that can precisely forecast the air quality for tomorrow, informing us of two important factors: the likelihood that the air will be unhealthy (the probability of it reaching a critical level) and how bad it will likely be(the predicted AQI number).