International Journal of Advances in Scientific Research and Engineering (IJASRE), ISSN:2454-8006, DOI: 10.31695/IJASRE
Not a member yet
1411 research outputs found
Sort by
Law Enforcement of Mineral and Coal Mining in the Banyuwangi Region, Indonesia
Full text linkThis study examines law enforcement against unauthorized sand mining activities in Wangkal Neighborhood, Kalipuro Subdistrict, Banyuwangi Regency. The issues discussed include the implementation of Article 158 in conjunction with Article 35 paragraphs (2) and (3) of Law Number 4 of 2009 on Mineral and Coal Mining, as amended by Law Number 2 of 2025, as well as the factors hindering law enforcement in the region. The findings indicate that law enforcement is still suboptimal, as evidenced by the low number of prosecuted cases and the generally light sentences imposed, suggesting a lack of deterrent effect on illegal mining perpetrators. Obstacles to law enforcement include limited personnel and capacity of investigators, insufficient infrastructure, complex investigative procedures, and low public legal awareness. This study recommends increasing the capacity of law enforcement officials, broader distribution of authority, and stronger legal education for the community to enhance the effectiveness of law enforcement. With more decisive and structured handling, it is expected that environmental damage caused by illegal mining can be minimized
Adoption of Government Electronic Payment Gateway (GePG): Adoption and Users Satisfaction Integrated Concept.
No full textDigital transformation has significantly impacted the financial sector by enhancing responsibility, efficiency, and openness, particularly through electronic payment systems. In Tanzania, the Government Electronic Payment Gateway (GePG) aims to expedite public financial operations, enhance revenue collection, and improve service delivery. This study explores user satisfaction with the adoption and usage of Tanzania’s Government Electronic Payment Gateway (GePG). The study combines three well-known theoretical models (Technology Acceptance Model (TAM), the Unified Theory of Acceptance and Use of Technology (UTAUT), and DeLone & McLean’s IS) to create a framework for evaluating user behavior, focusing on factors like ease of use, system quality, and external influences on adoption choices. Data were collected from 349 users across four regions of Tanzania using a quantitative survey methodology. Structural equation modelling (SEM) was used for analysis and testing the study hypothesis.
The results indicate that users' decision to continue using GePG is notably affected by factors like perceived usefulness, social influence, facilitating conditions, system quality, and information quality. Significantly, perceived usefulness emerged as the main factor, underscoring the role of system efficiency in promoting acceptance. However, the study revealed that continuous usage was not notably affected by perceived ease of use and service quality. Despite the presence of government e-payment systems, the survey shows that users prioritize functional benefits over ease of use.
This study fills the existing research gap of examining end-users’ acceptance and satisfaction of GePG. This study presents a validated conceptual model that is applicable to Tanzania and other developing countries, promoting ongoing technology adoption and efficient public service delivery
Using a pre-trained Network to recognize the “Kan group”
Full text linkOne of the most important fields of artificial intelligence is processing the Arabic language; its goal is to enable computers to understand and analyze texts written in human language. The arrival of pre-trained neural networks has made it feasible to enhance the precision of Arabic text analysis while identifying "kāna" and its sisters in sentences with greater efficiency. Therefore, in this work a framework for obtaining a qualitative identification of the “kāna” and its sisters in Arabic texts through the use of the pre-trained neural networks (DNNs) using the computed Zernike moments was developed. An extensive corpus of Arabic text examples that includes all instances of the use of "kāna" and its sisters in varying fonts and sizes was compiled. Arabic Zernike Moments were calculated to use as the base model using pre-trained DNNs. We first performed pre-training upon the collected dataset, then fine-tuning on the specific recognition task of “kāna” and its sister. Metrics such as accuracy, recall and F1 score are used to evaluate the models performance. The trained until now model detected well "kāna" and his sisters (those words behaved like "kāna") and got high accuracy on identifying these grammatical tools in Arabic texts. The results also reflect a good performance of the model in dealing with the variety of contexts in which kana and its sisters are found
Pattern Recognition for Fraud Detection in Mobile Money Transactions in Nigeria using a Stacked Ensemble Technique
Full text linkMobile money has made it easier for people in Nigeria to store, send, and receive fund using their mobile phones. This makes financial services more accessible to both rural and urban communities. Although this growth has improved financial inclusion, it has also created opportunities for fraud where existing detection systems struggles due to high false-positive rates that disrupt legitimate transactions. Our study explored a stacked ensemble machine learning model aimed at improving fraud detection while reducing false positives. We used the PaySim dataset from Kaggle, which originally contained 6,362,620 transactions. Random undersampling was used to handle class balance in the dataset, resulting in 13,140 records for model development. Exploratory analysis identified common fraud patterns, including transaction amount, frequency, and unusual balance changes. XGBoost, Random Forest, and Logistic Regression serve as base models, with LightGBM as the meta-learner. We evaluated performance using precision, recall, F1-score, false-positive rate (FPR), and AUC metrics. The model achieved a recall of 99.51% and an AUC of 99.4%, outperforming individual base models. Hyperparameter tuning reduced the FPR by 19%, from 0.94 to 0.76, reducing the misclassification of legitimate transactions. These findings revealed that a stacked ensemble approach detects fraud more effectively and reduces false positives and could be extended to other areas of financial fraud across Nigeria’s financial ecosystem
Methodological Knowledge Sharing Framework for a Natural Language Processing Chatbot: Review of dimensions
Full text linkKnowledge sharing is a concept that has been researched for many years. The success of knowledge sharing has been through good organization and consideration of different dimensions affecting the knowledge sharing process. While the majority of the knowledge-sharing dimensions are known, little has been written on the knowledge-sharing dimensions necessary for developing a knowledge-sharing framework that uses Natural Language Processing chatbot technology as the knowledge-sharing technique. The purpose of this article is to present dimensions and the proposed knowledge-sharing framework that use Natural Language Processing chatbot as the knowledge sharing technique. The systematic literature review was done by reviewing 28 peer-reviewed papers from 2014 to 2024 based on the inclusion and exclusion criteria set for the study. The findings revealed twelve (12) knowledge-sharing dimensions where nine (9) among them were selected to be used in the proposed framework. The selected dimensions are (1) knowledge domain, (2) knowledge actor, (3) source of knowledge, (4) knowledge collection, (5) knowledge processing, (6) chatbot creation, (7) knowledge use, (8) behavioural factors, and (9) knowledge owner. The study contributes to the body of knowledge by providing important dimensions for a chatbot development framework. In addition, the study contributes to practitioners by providing a proposed chatbot development framework that can be used during the Chabot development process. Furthermore, the results of this study will be beneficial to knowledge-sharing, and Natural Language Processing researchers.
Articles must include an Abstract of 250 words. The abstract should state briefly the purpose of the research, the principal results and major conclusions. The abstract should not repeat the information which is already present in the title. References should be avoided.
 
Gateway Energy Aware Protocol for LoRaWAN
Full text linkThe Long Range Wide Area Network (LoRaWAN) standard was primarily proposed to connect a large number of Internet of Things (IoT) devices over wide geographical areas while maintaining low energy consumption. In these networks, end devices and the network server exchange uplink and downlink messages through gateways. The selection of the gateway for downlink transmissions is often based solely on the received signal strength. However, this approach can lead to unbalanced energy consumption among gateways, thereby reducing the overall network lifetime—particularly in contexts such as precision agriculture or environmental monitoring in remote areas, where gateways cannot be powered by the electrical grid and their batteries are difficult to replace. To address this issue, we propose a communication protocol designed to extend the lifetime of LoRaWAN networks by considering the energy constraints of the gateways. This protocol integrates the residual energy of the gateways into the downlink selection process. Experimental results demonstrate that the proposed approach increases the network lifetime by approximately seven times compared to the standard protocol, while maintaining network performance at an acceptable level of the abstract
Recognizing Individual Faces of Variable Sizes using a Face Recognition Algorithm
Full text linkFace recognition that uses a person's face to identify or verify their identification is called face recognition. It functions by recognizing and quantifying face features in a picture. Technologies that use facial recognition can identify faces in photos or videos, identify if two faces are of the same person, or look for a particular face in a vast library of previously taken photos. When users enroll or log in, biometric security systems use facial recognition technology to uniquely identify each person. The objective of study is to detect and identify face in coloured images ( single or multiple faces) of human. A statistical algorithm based on skin color information, together with the facial features representing holes such as the eye or mouth, was applied. The study involved, among other things, the influence of the type of colored pictures employed in the test was examines, as human images from different races were used. The results confirmed this technology's success in identifying each person's individual face within the group
Development and Control of time-related interval for ambiguous knowledge bases (Type II) using genetic algorithms
Full text linkThe Interval Type-2 Fuzzy Logic Control (IT2FLC) utilizes a genetic algorithm (GA), known as the Genetics Interval Type-2 Fuzzy Network (GIT2FS), to optimize the fuzzy parameters, including fuzzy functions for membership and fuzzy regulation bases. After a brief discussion of the genetic fuzzy system GFS, the suggested design is described. Type reductions and defuzzification are included in the output processing of interval type-2 fuzzy logic circuits. Although researchers have recently developed numerous effective type reduction techniques, there are currently no practical plans to enhance the output of defuzzification. The kind of interval type-2 fuzzy set is reduced using the type reduction algorithm presented in this paper, which also produces the best defuzzified output from the type-reduced set. The planned type reduction is also carried out offline (in other words, the controller has been reduced to type-1 in practical applications). It greatly lowers the computational expense and makes it easier, actually, to develop controllers. Problems with truck backing control are used to show the viability of the suggested approach. The study showed that, in terms of speed, computational complexity, and resilience, the suggested technique performs better than typical IT2-FLCs
The Fixed-Implicit Object Alignment Method for Image Processing
Full text linkThe challenge facing image processing (IP) systems in scenes is the process of seamlessly integrating virtual objects with real scenes. There are several challenges, including the alignment problem, which is the inability to overcome the alignment problem due to the misalignment between virtual objects and their desired locations in real scenes, which leads to a complete deterioration of the system's performance, thus disrupting the image processing process to create an environment and weakening the user experience. Note that the alignment problem appears in the specified region or object, so we use an approach for alignment of fixed objects implicitly. Our experiment focuses on resolving discrepancies between the specified location of the virtual object and the corresponding real object, unlike dynamic alignment that continuously adapts to the entire scene. This work is to fix the virtual object in a fixed reference region within the real-world framework, which led to enhanced results and reduced cases of alignment failure in critical areas
Phenotypic characterization of the exotic Genotypes and widely cultivated common bean Genotypes in Southern Highlands of Tanzania
Full text linkCommon bean (Phaseolus vulgaris L.) is a key legume crop that is prized for its nutrition, adaptability, and ability to promote sustainable agriculture through nitrogen fixation. In Tanzania southern highlands, common bean characterization creates limitations to its potential yield and responsiveness to varying environments. In this research, phenotypic characterization of 12 popular common bean genotypes in southern highlands of Tanzania and imported genotypes from CIAT Colombia were studied. In the evaluation, promising genotypes with desirable phenotypic characteristics were identified. The genotype 22ACC02333 was characterized by strong growth and heavy leaf intensity, 22ACC03221 were characterized by medium-sized leaves and heavy curvature of pods, 22ACC02881 was characterized by pale green leaves with pigmented stems, and 22ACC02433 was characterized by spreading growth and twining habit. All the above genotypes shown favorable phenotypic characters of flowers, yielding, testa color, and heavy growing structure, indicating their suitability for future breeding programs. The study highlighted the market worth of such genotypes since they complied with market requirements for the uniformity of pod size, seed pigmentation, storage capacity, and yield potential, making them suitable for consumer and processing uses. Further, the phenotypic characteristics scored for these genotypes rate them with ability to withstand against biotic and abiotic stress which are vital for small scale farmers’ common bean production in the regions. The study is a strategic manual to breeding programs seeking to develop better productive, resistant and marketable improved varieties that are valued materials for viable agriculture development across Tanzania's southern highlands and other areas affected by climate variability.