Development of Neutrosophic Pareto Distribution for Survival Analysis

Ahmedia Musa M.

doi:https://doi.org/10.54216/IJNS.250426

Journal of Artificial Intelligence and Metaheuristics | 2025

Predictive Analysis of Groundwater Resources Using Random Forest Regression

The lack of water is one of the most crucial problems of our day; therefore, optimized water resource management and predictions gathered by patrons are of utmost importance. In the turmoil of a country like India, which lives a variety of lifestyles and has a complicated network of rivers, the urgent need for an active point of view to take care of water shortages becomes exceptionally vital. In this study, India’s groundwater, available at the district level for the year 2017, was the area of focus, with this analysis utilizing a dataset of 689 rows, each representing a district, and 16 columns describing the various groundwater extraction and recharge metrics. The study involves five regression models adapting RandomForestRegressor, DecisionTreeRegressor, MLPRegressor, KNeighborsRegressor, and SupportVectorRegression for water resource evaluation and prediction. Every model is appraised by using a thorough metrics set where we incorporate Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Explained Variance Score (EVS), Max Error, Median Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), R-squared (R2), among others. Our results put the spotlight on RandomForestRegressor, making MSE measures the same as 0.000206624, endorsing its better performance versus the criteria considered. The approach used in this model provides us with an ensemble effect that makes it more robust in the sense that we can capture the interrelationships within the dataset in a comprehensive way. DecisionTreeRegressor also provides nice options for precision and transparency. The use of such models depicts the potential value of predictive analytics, which has the role of improving resource management and planning because we can all agree that the impending water crisis is also a fact. These research outcomes provide us with important data for well-informed decisionmaking and strategic management of water reserves through all avenues and most affected areas to air most of the impact of water scarcity.  

groups

Khaled Sh. Gaber mail -

Manish Kumar Singla mail

link https://doi.org/10.54216/JAIM.090102

Volume & Issue

Vol. Volume 9 / Iss. Issue 1

Details open_in_new

Journal of Artificial Intelligence and Metaheuristics | 2025

Developing A Hybrid Machine Learning Algorithm for Anemia Diagnosis

The utilization of artificial intelligence (AI) algorithms has significantly transformed the field of blood disease diagnosis, enabling enhanced capabilities in prediction, categorization, and optimization. However, there is still a lack of research exploring the advancement of hybrid machine learning models that combine qualitative and quantitative datasets to address issues associated with blood diseases. To tackle this gap, we evaluate algorithmic combinations using datasets that include key characteristics from complete blood count (CBC) examinations. This manuscript presents an evaluation of prominent deep learning models, such as CNN, RNN, and RCNN, as part of our methodology. The assessment identified XGBoost as the optimal machine learning algorithm, and RCNN as the best deep learning model. Consequently, we propose a hybrid model named ‘RCNNX,’ which integrates Robust Scaler, SelectKBest feature selection, RCNN, and the XGBoost algorithm. The hybrid model, ‘RCNNX,’ achieves exceptional testing accuracy levels of 100% and 95.12% on the Anemia Diagnosis Dataset and a second dataset, respectively. Additionally, it demonstrates recall rates of 100% and 94.64% for the same datasets. These findings highlight the superiority of the proposed model, as it effectively utilizes feature selection to reduce the number of input variables, minimizing the risk of overfitting. Moreover, XGBoost enhances the predictive accuracy of RCNN.

groups

Safa S. Abdul-Jabbar mail -

Alaa k. Farhan mail -

Abdul Hafeez Kandhro mail

link https://doi.org/10.54216/JAIM.090103

Volume & Issue

Vol. Volume 9 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2025

Modelling Software Development Effort Using Data-Driven Models

Software effort estimation is highly significant for project management regarding the bidding process since underestimation leads to financial losses, while overestimation may bring the chance of losing a competitive bid. Whereas numerous models have been designed up until now, those developed upon machine learning, Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Artificial Neural Networks (ANN) have emerged as preeminent technologies. The proposed research will explore the effectiveness of using the ANN and ANFIS approaches in the estimation of effort for NASA datasets by 13 observations used for training and the rest for the test. To check the precision of models, several measures are used to evaluate the accuracy of the developed model, including the correlation coefficient, RMSE, and MMRE. The findings demonstrate that ANN and ANFIS exhibit superior performance, yielding much higher prediction accuracy compared to conventional Models including Walston-Felix, Doty, Bailey-Basili, and Halstead. It emphasizes ANN and ANFIS as reliable and straightforward software effort estimating methodologies, hence yielding significant enhancements in estimation precision and competitiveness. Their high performance underlines their usefulness to project managers who seek accurate predictions. This study strongly recommends the application of data-driven approaches like ANN and ANFIS to enhance the overall estimation accuracy in software project bidding.

groups

Zainab Rustum Mohsin mail -

Firoj Khan mail

link https://doi.org/10.54216/JISIoT.150203

Volume & Issue

Vol. Volume 15 / Iss. Issue 2

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2025

Deep Secure: An Integrated Approach to Anomaly Detection and Cryptographic Protection in Industrial Cyber-Physical Systems

Industrial Cyber-Physical System (CPS) signify a noteworthy development in industrial automation and control, combining physical and digital parts in order to improve the efficacy, trustworthiness, and functionality of numerous industrial procedures. Industrial CPS are helpful in a huge range of industries such as transportation, energy, manufacturing, and healthcare.  Intrusion detection systems (IDs) assist as vigilant protectors, constantly observing network and physical modules for any illegal access, variances, or doubtful actions. They deliver initial threat recognition and prevent safety breaks and operating troubles. In addition, cryptographic protection guarantees the privacy, honesty and genuineness of data that spread across Industrial CPS systems. By utilizing innovative encryption and authentication devices, cryptographic solutions defense complex data from capture or damage preserving consistency and confidentiality of dangerous industrial procedures. The combination of these safety actions creates a strong defence device, boosting the flexibility of Industrial CPS besides developing cyber threats and protecting the reliability of vital industrial processes. This article presents a Deep Secure: An Integrated Approach to Intrusion Detection and Cryptographic Protection in Industrial CPS environment. The proposed model aims to integrate intrusion detection and cryptographic-based secure communication protocol for industrial CPS environments. The Deep Secure model comprises two major phases: intrusion detection and secure communication. Primarily, the intrusion detection process comprises a self-attention-based bidirectional long short-term memory (SA-BiLSTM) technique. Besides, the deer hunting optimization algorithm (DHOA) achieve hyperparameter tuning of the SA-BiLSTM technique. Moreover, a secure communication protocol is designed by the use of the ElGamal cryptosystem. The experimental result of the Deep Secure method was tested in terms of dissimilar measures. A comprehensive result analysis highlighted the advanced performance of the Deep Secure method when associated to other current approaches.

groups

Sameer Nooh mail

link https://doi.org/10.54216/JISIoT.150204

Volume & Issue

Vol. Volume 15 / Iss. Issue 2

Details open_in_new

Journal of Artificial Intelligence and Metaheuristics | 2025

K-Nearest Neighbors Approach to Analyze and Predict Air Quality in Delhi

The study considers the community of ”urban air quality improvement in modern cities” using an extensive dataset obtained from ”Air quality data of Delhi, India” for the period between 25 November 2020 and 24 January 2023. Research aims to significantly reduce air pollutants, including particulate matter, including PM2.5 and PM10, NO2, SO2, CO2, O3, and others. Different machine learning models are being used for airquality level forecasts. Among the models assessed, the Nearest Neighbors algorithm comes out on top and exhibits a very low Mean Squared Error (MSE) of 0.0002. The model’s superb precision is further supported by very low statistics in other key metrics, which confirm the Nearest Neighbors approach to forecasting the quality of air in urban zones. The Nearest Neighbors algorithm is shown to have its place in the application as a tool in the hands of researchers and decision-makers pursuing the fight against air pollution is also a signal of its efficiency and broad applicability. This modeling approach has thus the potential to first identify and later pinpoint localized empirical patterns and statistical dependencies from the data set. Its high predictive precision makes it a very valuable assistant to public health and environmental management, especially so in regions like Delhi.

groups

Ahmed Mohamed Zaki mail -

Shahid Mahmood mail

link https://doi.org/10.54216/JAIM.090104

Volume & Issue

Vol. Volume 9 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2025

Explainable AI-Driven Gait Analysis Using Wearable Internet of Things (Wiot) and Human Activity Recognition

Due to the rapid expansion of the Internet of Things (IoT), supportive systems for healthcare have made significant advancements in both diagnosis and treatment processes. To provide optimal support in clinical settings and daily activities, these systems must accurately detect human movements. Real-time gait analysis plays a crucial role in developing advanced supportive systems. While machine learning and deep learning algorithms have significantly improved gait detection accuracy, many existing models primarily focus on enhancing detection accuracy, often neglecting computational overhead, which can affect real-time applicability. This paper proposes a novel hybrid combination of Sparse Gate Recurrent Units (SGRUs) and Devil Feared Feed Forward Networks (DFFFN) to effectively recognize human activities based on gait data. These data are gathered through Wearable Internet of Things (WIoT) devices. The SGRU and DFFFN networks extract spatio-temporal features for classification, enabling accurate gait recognition. Moreover, Explainable Artificial Intelligence (EAI) assesses the interoperability, scalability, and reliability of the proposed hybrid deep learning framework. Extensive experiments were conducted on real-time datasets and benchmark datasets, including WHU-Gait and OU-ISIR, to validate the algorithm’s efficacy against existing hybrid methods. SHAP models were also employed to evaluate feature importance and predict the degree of interoperability and robustness. The experimental results show that the method, combining Sparse GRUs and Tasmanian Devil Optimization (TDO)-inspired classifiers, achieves superior accuracy and computational efficiency compared to existing models. Tested on real-time and benchmark datasets, the model demonstrates significant potential for real-time healthcare applications, with an AUC of 0.988 on real-time data. These findings suggest that the approach offers practical benefits for improving gait recognition in clinical settings.

groups

Ponugoti Kalpana mail -

Sarangam Kodati mail -

L. Smitha mail -

Dhasaratham mail -

Nara Sreekanth mail -

Aseel Smerat mail -

Muhannad Akram Ahmad mail

link https://doi.org/10.54216/JISIoT.150205

Volume & Issue

Vol. Volume 15 / Iss. Issue 2

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2025

Ensemble of Machine Learning Model with Tuna Swarm Optimization-Driven Feature Selection for Cybersecurity Threat Detection and Classification Approach

The initial identification of cybersecurity events like attacks is challenging provided the continuously growing threat environment. Despite state-of-the-art surveillance, advanced attackers can apply for more than 100 days in a system before being detected. Guaranteeing cyber security is a composite task that depends on area of interest and needs cognitive capabilities to control possible threats from larger quantities of network data. The most important task of a cyber-security analyst is to safeguard a network from damage. Numerous technological developments in network and information security have enabled progressive monitoring and threat detection for the predictors, but the responsibilities they carried out could not be automated completely. Hence, in recent times’ Artificial intelligence (AI), mainly deep learning (DL) and machine learning (ML) algorithms, has been utilized to expand a beneficial data-driven intrusion detection system (IDS). Many standard ML classification methods provide intelligent facilities in the area of cyber-security, mainly for intrusion detection. This study develops a Tuna Swarm Optimization-Driven Feature Selection with Ensemble of Machine Learning Models for Cybersecurity Threat Detection and Classification (TSOFSEML-CTDC) technique. The proposed TSOFSEML-CTDC model concentrates on detecting and classifying intrusions on the network. Initially, the TSOFSEML-CTDC algorithm performs data preprocessing using min-max normalization to convert an input data into a beneficial format. Then, the feature selection process has been carried out using tuna swarm optimization (TSO) algorithm. For the classification of intrusion detection, ensemble of ML techniques was employed such as support vector regression (SVR) approach, least-square support vector machines (LSSVM) method, and modified extreme learning machine (MELM) technique.  At last, the hyperactive parameter optimization process is executed by using the coati optimization algorithm (COA). The experimental evaluation of the TSOFSEML-CTDC model occurs using a benchmark dataset. The stimulated results emphasized the enhanced performance of the TSOFSEML-CTDC method compared to existing approaches.

groups

K. Anitha mail -

K. Rajiv Gandhi mail

link https://doi.org/10.54216/JISIoT.150206

Volume & Issue

Vol. Volume 15 / Iss. Issue 2

Details open_in_new

Journal of Artificial Intelligence and Metaheuristics | 2025

Lung Cancer Prediction from Smoking Cause by Machine Learning Classification Models

The incidence of lung cancer varies in males and females, which occurs due to the abnormal and uncontrolled growth of cells in the lungs. It has a greater predilection in males as compared to females. Smoking is the most important risk factor for lung cancer. It causes serious breathing issues and also affects other organs. It increases the mortality rate both in young adults as well as in the older age group. Therefore, there is improvement in medical technologies to facilitate specialized diagnosis and treatment, but the mortality has not been controlled to a satisfactory extent. It is important to take preventive measures and precautions at the initial stages. Machine learning brings various advancements to the medical sector due to which various diseases can be detected at an early stage. In this paper, we presented different machine learning classifier techniques used for the classification of the present lung cancer data in the UCI machine learning repository as benign and malignant. The dataset is divided into cancerous and non-cancerous by converting the input data into binary form and using the classifier technique in theWeka tool. This specifically includes classifiers used: Logistic Regression, Random Forest, Support Vector Machine (SVM), k-Nearest Neighbors (k-NN), Decision Trees, and Na¨ıve Bayes. In addition, we study the effect of data preprocessing methods on our prediction accuracy, such as data normalization and feature selection. The study seeks to help develop various reliable resources for lung cancer identification, which are critical for diagnosing and treating patients in a timely manner and improving their outcomes.

groups

Nada M. Sallam mail -

P. K. Dutta mail

link https://doi.org/10.54216/JAIM.090105

Volume & Issue

Vol. Volume 9 / Iss. Issue 1

Details open_in_new

Journal of Artificial Intelligence and Metaheuristics | 2025

Comparative Analysis of Machine Learning Models for Daytime Power Generation Prediction

This paper proposes to evaluate how different machine learning techniques can be used to predict daytime power generation based on the ”Daily Power Generation Data” dataset. As a result of six models, which contain Random Forest Regressor, Decision Tree Regressor, Nearest Neighbors, Linear Regression, MLP Regressor, and SVR, a clear understanding has been accomplished by assessing the performance using multiple metrics. First of all, the Random Forest Regressor turned out to be the best in terms of the Mean Squared Error (MSE) of 3.57×10−6, which was the lowest among the three ML models. The introduction of the paper highlights the role of precise planning of the power market and the consecutive sections describing the topic mathematically. The table below, with a total list of performance issues, explains why the Random Forest Regressor is the superior full-proof model using the lowest MSE, highest explained variance, and great resistance to outlying samples. The paper thus gave various useful approval criteria that, to a great extent, we can choose the best model out of them because the Random Forest Regressor was in a position to get the highest performance metrics.

groups

Marwa M. Eid mail -

Anis Ben Ghorbal mail

link https://doi.org/10.54216/JAIM.090106

Volume & Issue

Vol. Volume 9 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Development of Neutrosophic Pareto Distribution for Survival Analysis

We provide a neutrosophic approach to the Pareto model, which is widely used to model survival data. In this paper, the neutrosophic Pareto model (NPM) is constructed under the framework of neutrosophic statistics, that can manage uncertain nature of data, commonly occur in many real word problems. This formulation generalizes the&ensp;classical model and is a useful method for dealing with fuzzy or uncertain data typically encountered in many applications in survival data. Using neutrosophic statistical framework, few key mathematic qualities of the proposed model such as its moments, survival function, and hazard rate are presented in the study. These properties are motivated and rigorously established to ensure theoretical soundness of the proposed model. Moreover, the maximum likelihood estimation (MLE) is used to estimate the neutrosophic parameters of the distribution. This approach is essential for deriving accurate parameter estimates from the data available, especially in cases where uncertainty or imprecision is present within the data as it is usually the case for any real-world situation. Based on the simulation experiment, we display the adequate performance of the suggested model. The simulations allow us to evaluate the performance of the routine as well as the stability of the model parameters across different settings. At the end, the real data analysis is conducted to show the applicability of proposed approach. The proposed model processes such a dataset filled with a range of uncertain values and presents its possibilities to be applied for information extraction from real world data sets that are abundant in uncertainty. Our results open a new avenue for neutrosophic statistical model approaches to the analysis&ensp;of survival data in subsequent studies.

groups

Ahmedia Musa M. Ibrahim mail

link https://doi.org/10.54216/IJNS.250426

Volume & Issue

Vol. Volume 25 / Iss. Issue 4

Details open_in_new