International Journal on Recent and Innovation Trends in Computing and Communication

Automate Everything: Mastering Salesforce Flows

2026-01-09T06:43:22+00:00

This work proposes the quantitative measures of Salesforce automation effect of three key CRM operations, which are quote approvals, rebate approvals, and customer course approvals. Findings obtained on a baseline of 90 days and a post-deployment of 90 days indicate that there have been significant changes in regard to the speed of the processes, their accuracy, and decrease in the number of man-hours performed manually. Flow Builder, Dynamic Forms, and Invocable Methods Apex along with each other helped to lessen the time of operation of the processes, reduce errors, and reduce the rate of rework of the thousands of transactions. The results affirm that the use of low-code automation allows generating quantifiable returns on performance and consistency. The evidence indicates that Salesforce Flows are very effective in enhancing reliability and efficiency in CRM business at the enterprise level.

AI-Driven Regulation-Aware BIM Framework for Clash Prioritisation and Digital Twin Integration in Saudi Vision 2030 Megaprojects

2026-02-06T07:15:20+00:00

Purpose
This study develops an AI-driven, regulation-aware framework that integrates Building Information Modelling (BIM) and the Saudi Building Code (SBC) to prioritise critical clashes and enable digital twin integration within Vision 2030 megaprojects.

Design/methodology/approach
A hybrid ensemble combining Random Forest (RF), Convolutional Neural Network (CNN), Graph Convolutional Network (GCN), and Graph Attention Network (GAT) was trained using hierarchical graph processing. SBC clauses were encoded into IFC features, with calibrated probabilities and a fixed cost-derived decision threshold.

Findings
Nested leave-one-project-out (LOPO) testing across five industrial federations demonstrated consistent improvements in AUROC, AUPRC, and calibration. The framework reduced coordination time by approximately 65% compared with incumbent workflows, with statistically significant results and large effect sizes.

Originality/value
This paper presents one of the first regulation-aware AI models for BIM clash prioritisation under the Saudi Building Code. The openly released framework enables reproducibility and provides a foundation for real-time digital twins in Saudi Vision 2030 projects.

Machine Learning Models for Alzheimer’s Disease Prediction: A Comparative Study

2026-02-13T08:11:36+00:00

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that significantly impacts cognitive and functional abilities. Early detection and accurate prediction of AD progression are critical for effective intervention and management. This study explores the predictive potential of machine learning algorithms, including Linear Regression, Multilayer Perceptron, SMOreg, Random Forest, Random Tree, and REP Tree, applied to Alzheimer’s-related datasets. The dataset comprises features such as demographic (ID, gender, handedness, age, education, socioeconomic status), cognitive (MMSE, CDR), and structural (ETIV, NWBV, ASF) attributes. Comprehensive analysis reveals the strengths and limitations of each model in handling the diverse dataset characteristics. The results demonstrate that tree-based methods like Random Forest and REP Tree provide superior accuracy, while neural network-based approaches like Multilayer Perceptron effectively capture nonlinear relationships. This research underscores the importance of integrating cognitive and structural metrics to enhance predictive capabilities, offering valuable insights for early diagnosis and personalized care strategies in Alzheimer’s disease.

Forensic Examination of Original and Scanned Signatures to Study Different Characteristics Using Image Processing Tool.

2026-02-13T10:00:16+00:00

There has been a simultaneous rise in the employment of image processing software applications in criminal operations to match the recent exponential surge in their use. However, for more effective processing and better storage, the majority of documents these days are created entirely online. The development of computer-based image processing tools and contemporary replication processes has simplified the process of digitally fabricating signatures. The goal of the current study is to identify the various characteristics of scanned (at varying dpi or resolutions) and printed (using inkjet and laser printers) signatures. Many printed signature samples were created from genuine documents, using traditional and digital tools like Adobe Photoshop. Identified traits include alignment errors, irregularities in half-tone patterns, variations in letter spacing, and over-deposition of ink particles. It could be possible to differentiate a scanned signature from the typical signatures of the real person by identifying and analyzing such traits. This study highlights the lack of research on offline scanned documents and makes the case for using the most recent image processing techniques to do forensic analysis on a variety of document types.

Survival Study on Student Academic Performance Classification based on Mental Health

2026-04-16T06:13:03+00:00

Educational Data Mining (EDM) is used to extract the important information from educational data. EDM identifies the trends from educational data to enhance the student academic performance. EDM uses the machine learning conceptsto recognize the learning, to improve teaching and to optimize the educational systems. Mental health issues are prevalent among students. Depression has significant obstacle for performing the long-term learning in educational system. Student dropout prediction is an important event for educational institutions and policymakers around world. Early student academic performance prediction is an essential research topic in educational data mining. Different deep learning and artificial intelligence methods are introduced to forecast the student academic performance. However, the existing prediction techniques failed to handle the student mental health and their mood changes. In order to address the existing issues, different artificial intelligence and deep learning methods is introduced for student academic performance classification based on mental health.

Leveraging Artificial Intelligence in Configure-Price-Quote (CPQ) Systems for Healthcare Manufacturing

2026-04-27T09:41:37+00:00

Healthcare manufacturing operates under stringent regulatory requirements, high product complexity, and increasing demand for customer-specific configurations. Configure-Price-Quote (CPQ) systems translate customer intent into manufacturable and compliant offerings, yet they often struggle at scale when relying solely on static rules and manually maintained knowledge bases. This paper presents an Artificial Intelligence (AI) enabled CPQ approach for healthcare manufacturing that combines knowledge-based configuration with governed, data-driven intelligence. We propose a validation-aware reference architecture and a digital-thread integration pattern that connects CPQ outputs with downstream PLM/ERP/quality systems to preserve traceability. Practical use cases are discussed: configuration recommendation, pricing and contract guidance, conversational access to catalog and release content, and compliance-oriented explainability, along with governance controls required for regulated environments.

Autonomous Solar Panel Cleaning Mechanism Powered by Solar Energy

2026-05-01T08:02:55+00:00

The efficiency of solar panels decreases over time due to the accumulation of dust, bird droppings, and environmental pollutants on their surfaces. This buildup reduces light absorption, thereby lowering power generation efficiency. Traditional cleaning methods are manual, labour-intensive, and often require external energy and water sources. To overcome these limitations, this project introduces a self-sustained, autonomous solar panel cleaning system powered entirely by the solar panel itself.

A 15W solar panel supplies energy to an ESP32 microcontroller, which intelligently manages a mini water pump and DC wiper motor for automated cleaning operations. The system stores surplus energy in a rechargeable lithium-ion battery, allowing it to function even under low sunlight conditions. To ensure efficiency, LDR (Light Dependent Resistor) or dust sensors monitor the panel’s surface cleanliness and automatically trigger cleaning cycles when required. The ESP32 also incorporates low-power deep sleep mode and supports Wi-Fi-based remote monitoring and scheduling.

This innovative design provides a cost-effective, energy-efficient, and fully autonomous solution for maintaining solar panel performance without relying on external power. It significantly reduces manual intervention, conserves water, and ensures consistent power output, making it suitable for both residential and industrial solar energy systems.

Effect of Seasons on Millimeter Wave (35 GHz) Prevailing in Foliage Depth of Desert Region of India

2026-05-05T09:27:05+00:00

Performance of millimeter wave is of interest to communication scientists, researchers, industrialist as it shown communication window with wide spectrum and fewer losses. Wavelength above 10 GHz is subject to molecular absorption which limits its uses for many purposes. In this paper effect of changing seasons is presented on basis of experimental results. A trans-receiver system of 35 GHz is used to quantify the effect of increasing foliage depth (Trees in line) with changing seasons. Molecular absorption of water molecules can be clearly pointed out in observations. It is observed that least attenuation occurred in autumn while spring offers maximum attenuation to signal due to water molecules. Non linear attenuation with increase in foliage depth also indicates phenomenon of coherent scattering of millimeter wave. Coherent scattering re-combines the scattered components which are in-phase. Total 35.16% of signal in autumn seasons is said to be coherently scattered while least coherency of 18.63 % is observed in summers. In winter and spring coherency of signal is observed as 25.88% and 23.33% respectively. Observations and calculations presented in the paper may help scientist to develop an algorithm for communication system which can work proper with varying seasons

Interpretable Wavelet Transforms: A Unified Framework for Frequency-Aware Learning and Dynamical System Identification

2026-05-09T05:43:04+00:00

Wavelet transforms provide simultaneous time–frequency localisation through a mathematically rigorous multi-resolution framework, yet their classical formulations fix filter coefficients independently of any learning objective. This paper makes three original contributions. First, we provide a unified mathematical treatment of learnable wavelet decomposition, establishing precise conditions under which trainable filters retain or forfeit perfect reconstruction guarantees. Second, we derive a gradient-based layer importance metric that quantifies which frequency bands drive model decisions, and demonstrate its application to physiological signal classification with reproducible experimental details. Third, we show that the multi-resolution signal decomposition principle underlying wavelets can serve as a structural prior for governing equation discovery in complex network dynamics, creating an explicit bridge between classical wavelet theory and modern neural symbolic regression. Worked examples on the ECGFiveDays benchmark and SIS epidemic dynamics illustrate the unified framework.

Digital HR Innovation and Employee Well-being in Moroccan Hotels

2026-05-14T14:37:06+00:00

Employee well-being is the most essential factor that can greatly influence the performance of any organization, and it is required for hotels in the tourism and hospitality industry. As part of human resource (HR) strategies in the modern workplace, technology has been extensively employed to improve employee’s well – being. The study revealed the significance of the digital HR innovation for the well-being of employees, serving in the hotels of Agadir in Morocco. A mixed-method approach was employed for this study. It consists of a quantitative survey of 100 hotel employees from 3 hotels and a qualitative literature review located in Agadir, Morocco. The employees' digital HR perceptions, including the integration, daily usage, workload impact, skill, and overall well - being of, were assessed with the help of the structured questionnaire having a 5-point Likert scale. The quantitative findings based on data’s descriptive statistics revealed employees' "cautious optimism" attitude at the workplace and significant reasons behind the stress, fear, and anxiety because of the fear of being monitored through digital surveillance. The survey respondents believed that the digital HR tools usage resulted in the lessening of administrative burdens and the increase in job satisfaction. The research findings revealed that a human-centered strategic framework that not only guarantees ethical governance but also recognizes technology as a means that supports and facilitates work and allows continuous assistance to improve employee’s well-being, required for hotels in Agadir, Morocco, for the effective utilization of Digital HR innovations.

An Intelligent Edge-AIoT Framework for Real-Time COVID-19 Safety Compliance: Integrating Deep Learning, RFID Authentication, and Precision Thermal Sensing

2026-05-16T06:40:42+00:00

Existing IoT-based access control systems for COVID safety, such as the referenced “Covid Safety Guidelines Detection” prototype, suffer from high inference latency, coarse ambient temperature sensing, and limited scalability. In this work, we propose an intelligent IoT-based autonomous access-control framework that overcomes these gaps through real-time deep learning, edge computing, and precise thermal sensing. The system integrates a camera for mask detection, an RFID reader for user authentication, and a contactless IR temperature sensor for fever screening. We develop a multi-stage decision model D(R,C,T) combining RFID validity (R), mask-classification confidence (C), and measured temperature (T) against thresholds, with D=1 granting entry only if all conditions are met. A convolutional neural network (CNN) is optimized for mask vs. no-mask classification with cross-entropy loss On-device (edge) processing on a Raspberry Pi 4 reduces round-trip delay. In experiments with 8,000 images (50% masked), our system achieves >98% mask-detection accuracy, 0.95 precision, and a mean inference latency of ~50?ms – 3× faster than a cloud-based baseline. Thermal sensing with an MLX90614 IR sensor attains ±0.5°C accuracy. By integrating real-time CNN inference, RFID authentication, and accurate body-temperature validation, our framework significantly improves upon prior work with rigorous mathematical modeling and quantitative validation. Key outcomes include high detection accuracy (>96%), low end-to-end latency (~60?ms), and robust real-world performance, demonstrating the feasibility of a scalable smart surveillance gateway.

A Comprehensive Review of Fish Disease Detection Systems Using Machine Learning and Deep Learning Techniques

2026-05-16T10:56:56+00:00

Fish diseases significantly impact aquaculture productivity, causing substantial economic losses and threatening food security. Traditional diagnostic methods are time-consuming, labor-intensive, and unsuitable for real-time monitoring. Recent advances in artificial intelligence, particularly machine learning and deep learning, have enabled automated fish disease detection systems based on image analysis. However, most existing approaches rely solely on visual data and overlook environmental factors such as water quality, which play a crucial role in disease development. This paper presents a concise review of fish disease detection techniques, including machine learning, deep learning, and IoT-based monitoring systems. It identifies key limitations such as lack of multimodal integration and limited real-time applicability. To address these challenges, a novel Multimodal Explainable Fish Disease Detection Network (MEFD-Net) is proposed, integrating image data, sensor data, and temporal modeling. The proposed approach improves accuracy, robustness, and interpretability, making it suitable for smart aquaculture systems.