International Journal of Systematic Innovation

A novel Ensemble Learning Model for Text Summarization with Improved Multilayer Extreme Learning Machine Auto Encoder (MLELM-AE) Using Machine Learning Algorithms

Sunil Upadhyay — 2025-10-16

The gigantic amount of electronic data gathered and analyzed has contributed to useful information sources that human beings need to manage easily. To make important decisions very quickly, the Automatic Text Summarization (ATS) method enables users to get relevance and expertise in a very small amount of time. ATS systems are notably extractive and abstractive, or by combining these two approaches, it is also used as a hybrid. The extractive technique involves extracting the most important sentences from the input document(s), then assembling these sentences to produce the summary. In the abstractive technique, a summary is produced by creating new sentences instead of picking sentences to convey the meaning of input text. A hybrid method intermingles of Abstractive and Extractive methods. In spite of many suggested techniques, the created summaries still don’t convey the actual meaning of text as compared to the man-made summaries. In this paper, the authors reviewed different techniques of text summarization and identified that the present models are computationally expensive and low training speed. To address this problem authors proposed an improved Multilayer Extreme Learning Machine Auto Encoder (MLELM-AE) and an Ensemble Learning Model for Text Summarization Using machine learning algorithms to enhance the quality of the summary. The proposed model is an ensemble of four models, i.e., Sentence2BERT (S2B), Auto Encoder (AE), Variational Auto Encoder (VAE), and Improved MLELM-AE. The performance of this in respect of execution time is 50015(ms) and ROUGH1 score of ensemble model is 0.865195. Thus, the results demonstrated that the recommended model is much better for text summarization as compared to existing models.

Exploring Satisfaction with Military Catering Services Using the Service Quality Model and Importance-Performance Analysis

yawen chan — 2025-10-16

The importance of military catering in military organizations cannot be overlooked, as it not only impacts the health and physical fitness of service members but also directly affects combat readiness and morale. This study focuses on a northern air force base, using the PZB (Parasuraman-Zeithaml-Berry) service quality model’s Gap 1 and Gap 5 as its framework. The aim is to investigate the perception gaps in catering service quality between catering personnel and meal users. An IPA (Importance-Performance Analysis) matrix is employed to further analyze the findings. The analysis reveals that, regarding "catering service quality," catering personnel who are actively serving without formal food service certification, and those with high school or college education, tend to place more emphasis on tangibility, reliability, empathy, and responsiveness. For service quality expectations, meal users who possess a college education and have obtained a food service certification show higher expectations in tangibility and reliability dimensions. Younger meal users, aged 18-25, who are uncertified and less experienced, report greater satisfaction with the catering service’s reliability, responsiveness, and assurance dimensions after their actual experience with the base’s services.

Regarding the perception difference in Gap 1 of the PZB model, the study suggests that services should prioritize user experience and ensure transparency by publicizing findings from meal review meetings. Feedback can be gathered through a satisfaction mailbox to address and efficiently amend any service deficiencies. For Gap 5 in terms of actual experience, meal users show particular concern for food safety measures and overall service quality, indicating that these areas should be maintained or enhanced. Regular training is recommended to improve the knowledge and effectiveness of catering personnel in these critical aspects.

A Graphics Processing Unit-Based Parallel Simplified Swarm Optimization Algorithm for Enhanced Performance and Precision

Wenbo Zhu — 2025-10-16

Graphics processing units (GPUs) have emerged as powerful platforms for parallel computing, enabling personal computers to solve complex optimization tasks effectively. Although swarm intelligence algorithms (SIAs) naturally lend themselves to parallelization, a GPU-based implementation of the Simplified Swarm Optimization (SSO) algorithm has not been reported in the literature. This paper introduces a CUDA Simplified Swarm Optimization (CUDA-SSO) algorithm on the CUDA platform, with a time-complexity analysis of O(N_gen ´ N_sol ´ N_var), where tt is the number of iterations, N_sol is the population size (i.e., number of fitness function evaluations), and N_var represents the required pairwise comparisons. By eliminating resource preemption of personal best (pBests) and global best (gBest) updates, CUDA-SSO significantly reduces the overall complexity and avoids concurrency conflicts. Numerical experiments demonstrate that the proposed approach achieves an order-of-magnitude improvement in run time with superior solution precision relative to CPU-based SSO, making it a compelling methodology for large-scale, data-parallel optimization tasks.

The An Adaptive Hybrid Clustering Framework for High-Precision Microarray Image Segmentation Using GA and BEMD

Ravikumar ch — 2025-10-16

The development of microarray technology has facilitated expression profiling analysis for various medical and agricultural research areas. Despite the improving range of applications, the precision in isolating microarray images remains a challenge due to noise and variances in spot morphology. This research proposes a hybrid and adaptive clustering solution that offers great improvement in terms of accuracy, segmentation, noise reduction, processing time, and overall efficiency. We use an adaptive K-means clustering approach enhanced with Genetic Algorithms (GA) and Bi-Dimensional Empirical Mode Decomposition (BEMD). The average segmentation accuracy tested by us was close to 95% with our method whereas the K-means methods averaged at 85%, showing how vastly superior our method is. Our methods with the added BEMD and GA achieved unparalleled cutting down noise by 80% and increasing SNR 200%, reducing overall efficiency. Through optimising the algorithm’s performance, our algorithms resulted in an average of 1.2 seconds for image processing time which showcases the efficiency of the algorithm. These results enable uniquely resolving complex problems in microarray image analysis, unlocking new solutions critical for gene profiling medicine and agriculture, enabling transformative advancements in the sectors.

Advanced Fault Detection in WSNs: A Metaheuristic-Driven Deep Learning Approach to Enhance Quality of Service

Gayathri R — 2025-10-22

Wireless Sensor Networks (WSNs) face significant challenges in fault detection, which directly impacts the Quality of Service (QoS) in dynamic environments. This research presents a novel framework to address these challenges, integrating a Dynamic Noise Filtering (DNF) technique with adaptive thresholding for effective noise removal while maintaining critical data integrity. The Rank-Based Whale Optimization Algorithm (RWOA) is employed for feature selection, optimizing model performance, and minimizing computational complexity. The core of the framework, the Hierarchical Attention-Based Deep Learning (HADL) model, leverages temporal convolutional layers, LSTM units, and hierarchical attention mechanisms to capture both short-term and long-term dependencies, resulting in exceptional fault detection accuracy. The proposed method demonstrates outstanding performance on the WSN-DS dataset, achieving precision, recall, F1 scores, and an AUC of 0.99 or higher for all fault classes. Comparative analysis reveals the superior performance of the framework in terms of accuracy, sensitivity, specificity, and computational efficiency. The approach not only improves fault detection but also enhances network reliability, reduces false alarms, and extends the operational lifespan of WSNs. This research offers a scalable solution for mission-critical applications, such as healthcare, environmental monitoring, and industrial automation, with potential for further enhancement through real-time deployment, multi-modal datasets, and hybrid optimization techniques.