Online reviews play a crucial role in shaping consumer decisions, especially in the context of e-commerce. However, the quality and reliability of these reviews can vary significantly. Some reviews contain misleading or unhelpful information, such as advertisements, fake content, or irrelevant details. These issues pose significant challenges for recommendation systems, which rely on user-generated reviews to provide personalized suggestions. This article introduces a recommendation system based on a Passer Learning Optimization-enhanced Bidirectional Long Short-Term Memory network classifier applicable to e-commerce recommendation systems with improved accuracy and efficiency compared to state-of-the-art models. More specifically, the proposed model achieves a high accuracy of 98.03%, F1 score of 98.03%, precision of 98.49%, recall of 97.57%, and minimum mean square error of 1.97 based on training percentage using the patio lawn garden dataset. These results, made possible by advanced graph embedding for effective knowledge extraction and fine-tuning of classifier parameters by the hybrid PLO algorithm, establish the suitability of the proposed PLO-Bi-LSTM model in various e-commerce environments. Received: 9 April 2025 | Revised: 26 June 2025 | Accepted: 24 August 2025 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The Amazon product data that support the findings of this study are openly available at https://cseweb.ucsd.edu/~jmcauley/datasets/amazon/links.html. Author Contribution Statement Hemn Barzan Abdalla: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition. Mehdi Gheisari: Conceptualization, Software, Data curation. Awder Ahmed: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Visualization. Bahtiyar Mehmed: Formal analysis, Investigation, Resources. Maryam Cheraghy: Validation, Resources, Writing – review & editing. Yang Liu: Validation, Investigation, Resources.

In an e-learning platform, information retrieval plays an enormous role through efficient processing. Recently, the education sector has increased its trend in online learning systems by generating a large amount of educational content based on student’s criteria. For this sophisticated data analysis scheme, several methods have been employed in recent studies; however, they have suffered from various limitations, including reliability issues, security problems, unauthorized disclosure of data, cost consumption, and interpretability challenges. To tackle these issues, a proposed framework, named the war strategy optimization-based bidirectional long short-term memory (WSO-BiLSTM) model, is designed in this research to reduce sensitivity to local optima and improve convergence stability, thereby achieving robust retrieval performance. With this perspective, the BiLSTM model captures the semantic information of documents in a dual direction for effective retrieval outcomes. Moreover, the model’s key features are extracted effectively by various feature extraction methods. The dynamic movement towards the optimal solution of the WSO algorithm enables the proposed model to retrieve the information more accurately in the information retrieval system. Experiments on an e-learning dataset show that, with a 90% training split, the proposed method achieves 97.90% accuracy, 98.45% precision, 97.90% F1-score, and 97.35% recall.

ML (mobile learning) has extended e-learning to a new paradigm of “anywhere, anytime learning”. The use of 3G and 4G Phones (high-speed data transfer) will be represented as a great opportunity for both learners and teachers to meet together, access and exchange information in virtual spaces whilst on the move. The aim of this work is to design a mobile learning framework for Students in higher education. Thus, this will provide better quality of teaching. Additionally, Mobile learning (M-learning) has turned into a critical instructive innovation part in advanced education. M-learning makes it workable for students to learn, team up, and impart thoughts among each other as much as web innovation and improvements will allow. In any case, M-learning acknowledgment by learners and instructors is basic to the occupations of M-learning frameworks. Attitudes towards M-learning innovation is an imperative method to measuring whether or not learners and instructors are prepared to utilize M-learning. Such attitudes will serve to distinguish qualities and shortcomings and encourage the advancement of the innovation foundation. We will investigate students and instructors' state of the arts in using M-learning in higher institutes for some of Private and Public universities of Ministry of higher education. As a result of student feedback regarding M-learning methods of teaching, a suitable framework for M-learning is proposed by reviewing many other frameworks and also by the analysis of results of a survey that asked many students and staff in higher education fields.

Accurate and comprehensive road extraction plays a crucial role in emergency response, street navigation, and urban planning. With the development of Remote Sensing (RS) technology, a large number of high-resolution RS images have been made available. Road extraction using RS images involves identifying and segmenting road networks from satellite or aerial images. However, managing occlusion and complex urban environments often leads to inaccurate road network extraction as occlusions such as trees, vehicles, and obstructed road features make it challenging to distinguish roads from other objects. Hence, this research introduces the Dense Atrous Pyramid Scene Parsing Network (DAPSPNet) for road extraction using RS images. The Dense Atrous Spatial Pyramid Pooling Module (DASPPM) is employed with varied dilation rates to capture multi-scale contextual information without compromising spatial details, thereby enabling effective modeling of extended roadways of varied widths. In the traditional Pyramid Scene Parsing Network (PSPNet), the DASPPM, a modification referred to as Dense Atrous (DA) is incorporated. When compared to existing methods, such as deep Convolutional Neural Networks (CNN), the proposed DAPSPNet achieves a precision of 93.09% and 91.08% on the DeepGlobe and Massachusetts datasets, respectively.