- All articles 732
-
International Journal of Information Engineering and Electronic Business @ijieeb
Journal articles - International Journal of Information Engineering and Electronic Business
All articles: 732
Scientific article
The aim of this paper is to investigate the performance of time delay neural networks (TDNNs) and the probabilistic neural networks (PNNs) trained with nonlinear features (Lyapunov exponents and Entropy) on electroencephalogram signals (EEG) in a specific pathological state. For this purpose, two types of EEG signals (normal and partial epilepsy) are analyzed. To evaluate the performance of the classifiers, mean square error (MSE) and elapsed time of each classifier are examined. The results show that TDNN with 12 neurons in hidden layer result in a lower MSE with the training time of about 19.69 second. According to the results, when the sigma values are lower than 0.56, the best performance in the proposed probabilistic neural network structure is achieved. The results of present study show that applying the nonlinear features to train these networks can serve as useful tool in classifying of the EEG signals.
Free
Classification of Multilingual Financial Tweets Using an Ensemble Approach Driven by Transformers
Scientific article
There is a growing interest in multilingual tweet analysis through advanced deep learning techniques. Identifying the sentiments of Twitter (currently known as X) users during the IPO (Initial Public Offering) is an important application area in the financial domain. The number of research works in this domain is less. In this paper, we introduced a multilingual dataset entitled as LIC IPO dataset. This work also offers a modified majority voting-based ensemble technique in addition to our proposed dataset. This test-time ensembling technique is driven by fine-tuning of state-of-the-art transformer-based pretrained language models used in multilingual natural language processing (NLP) research. Our technique has been employed to perform sentiment analysis over LIC IPO dataset. Performance evaluation of our technique along with five transformer-based multilingual NLP models over this dataset has been reported in this paper. These five models are namely a) Bernice, b) TwHIN-BERT, c) MuRIL, d) mBERT, and e) XLM-RoBERTa. It is found that our test-time ensemble technique solves this multi-class sentiment classification problem defined over the proposed dataset in a better way as compared to individual transformer models. Encouraging experimental outcomes confirms the efficacy of the proposed approach.
Free
Classification of the User's Intent Detection in E-commerce systems – Survey and Recommendations
Scientific article
The personalized experience gets more and more attention these days. Many e-commerce businesses are looking for methods to deliver personalized service. Consumers are expecting, if not demanding, highly personalized experiences. Moreover, customers are typically willing to spend more when they receive such a custom-tailored service. A prerequisite to provide a genuinely personalized experience is to understand the customer. Intent detection is a new and challenging approach in modern e-commerce to understand the customer. We find that various aspects of customer intent detection can be tackled by leveraging tremendous recent recommendation systems' progress. In this work, we review existing works from different domains that can be re-used for customer intent detection in the e-commerce. Even though many methods are used, there is no comparison of available approaches. Based on a review of nearly 100 articles from 2015 until 2019, we propose a categorization of types of intent detection, personalization context, building a customer profile, and dynamic changes in user interests handling. We also summarize existing methods from applicability in the e-commerce domain, including the aspect of the General Data Protection Regulation requirements. The paper aims at the classification of applied techniques and highlights their advantages and disadvantages.
Free
Classifying IoT Device’s Traffic Traces Using Network Traffic Characteristics
Scientific article
The escalating proliferation of devices, including both Internet of Things (IoT) and non-IoT devices, has triggered a suite of emergent security challenges in cyberspace, such as accurate device identification and authentication. The wide array of device types, protocols, and usability exacerbates these challenges. While conventional addressing schemes such as the logical Internet Protocol addressing and physical Media Access Control addressing schemes are integral for communication, they are susceptible to spoofing attacks. Device fingerprinting can be used to address the issue of identifying devices and traffic types using only implicit identifiers such as network traffic characteristics. In this paper, supervised machine learning based a device fingerprinting model has been proposed for the classification of both IoT and non-IoT devices on three levels based on their communication traffic characteristics. A meticulous feature selection process, employing two attribute evaluators, identified a subset of twenty features crucial for generating unique fingerprints from a large set of features pool. Three publicly available datasets and two supervised classifiers were utilized for evaluation purposes. Experimental results illustrated that the proposed model attained a classification accuracy exceeding 99% in discerning between known and unknown traffic traces (Level-1) on both the UNSW IoT and D-Link IoT datasets using the Random Forest (RF) classifier, and 99.74% accuracy in classifying network traffic types (Level-2) on the UNSW dataset. Individual device identification (Level-3) proves equally robust, with the RF and J48 classifiers achieving 99.03% and 98.14% accuracies on the UNSW non-IoT and IoT datasets, respectively. These findings underscore the potential of the device fingerprinting model in enhancing network security. The model’s robust classification capabilities across various datasets and identification levels make it a valuable asset in tackling modern security challenges in networked environments.
Free
Classifying Similarity and Defect Fabric Textures based on GLCM and Binary Pattern Schemes
Scientific article
Textures are one of the basic features in visual searching,computational vision and also a general property of any surface having ambiguity. This paper presents a texture classification system which has high tolerance against illumination variation. A Gray Level Co-occurrence Matrix (GLCM) and binary pattern based automated similarity identification and defect detection model is presented. Different features are calculated from both GLCM and binary patterns (LBP, LLBP, and SLBP). Then a new rotation-invariant, scale invariant steerable decomposition filter is applied to filter the four orientation sub bands of the image. The experimental results are evaluated and a comparative analysis has been performed for the four different feature types. Finally the texture is classified by different classifiers (PNN, K-NN and SVM) and the classification performance of each classifier is compared. The experimental results have shown that the proposed method produces more accuracy and better classification accuracy over other methods.
Free
Client Server iOS in Player versus Player (PVP) of “Borneo Snap”
Scientific article
Borneo Snap is a Kalimantan’s animals snap card game. Play proceeds with the players taking it in turns to remove a card from the top of their deck and place it face-up on a central pile. If two cards placed consecutively on the pile are identical (same picture) then the first player to shout "Snap!"will get all of it. This game is built for the iOS platform. Game Player Versus Player (PVP) Borneo Snap using peer-to-peer API from Game Kit (XCode Framework) and Wi-Fi or Bluetooth, but actually, Borneo Snap uses client-server architecture model, each player in a player versus player game session only can communicate with server intermediaries. If the player sends data updates when playing cards to other players, this data update will first be through the server, then forwarded to all other players. The result of this research, with client-server framework Borneo Snap can be played by more than 1 player and more iOS gadget too
Free
Scientific article
The goal of the question-answering system is to respond to user queries expressed in natural language. Unlike search engines, the closed domain question answering systems are specialized to specific domains, providing concise and precise answers often derived from structured data. This paper focuses on a question-answering system tailored for crime events, capable of addressing both statistical and contextual inquiries. In terms of crime statistics, the fine-tuned GPT-3 model outperforms the USE, TAPAS, TAPEX, and GPT-3 models, while for context-based crime-related queries, the fine-tuned RoBERTa model surpasses the BERT and RoBERTa models. This system is capable of providing the responses in natural language format, supplemented with relevant data visualizations. The models are train on Q2A and NewsQA datasets while it is tested on NCRB and NewsTimes datasets. The Q2A and NCRB datasets are used for statistical queries while NewsQA and NewsTimes datasets are used for contextual inquiries. The paper presents an analysis of various models and showcases results for sample case studies. Such a system can prove valuable in applications where users seek to study criminal cases or gather pertinent insights for specific cases. Furthermore, it can assist in understanding patterns and trends in criminal events, particularly concerning geospatial information. Linking crime event-based question-answering systems to geospatial information facilitates exploration of niche areas and furnishes precise details about local crime with minimal hype and hence worth exploring.
Free
Cloud and On-premises Based Security Solution for Industrial IoT
Scientific article
In this paper, we take Industrial IoT (IloT) as a main point, where we touch on the direction of Industrial IoT concepts and connectivity protocols used by Industrial IoT devices. Moreover, we go into deep security challenges the Industrial ecosystem faces. Nowadays, most industries focus on specific protocols in their smart IoT devices. In return, we mainly focus on Message Queuing Telemetry Transport (MQTT) protocol, MQTT server where IoT devices are connected to it, and secure connectivity among server, cloud, and end user. Our purpose here is to describe the security approach for server and cloud-based environments and the utilization of cloud security tools such as IoT-hub, Network Security Group (NSG) and virtual private network (VPN). In more detail, here we have indicated proposed solution by separating into device, on-premises and cloud zone sections, proper technologies which are being used in modern security approaches and comparison with traditional IoT security approaches. This article enables readers to obtain fundamental knowledges on available technologies which are utilized in industrial areas and real-time scenarios where this solution is deployed.
Free
Coalition Tolerance to Multiple Failures of Its Members
Scientific article
The paper deals with coalitions whose members are unselfish. Coalition members do their best to complete the arising tasks, and do not expect to receive a reward. A coalition member can be an entity such as a social or governmental organization, a military unit, or a complex technical device such as an autonomous robot, or any other entity that has the capabilities, willingness, and ability to cooperate. The paper considers the non-redundant coalitions, which have only those coalition members without whom they cannot perform the tasks. In the paper, we consider only the situations when substitution of failed coalition members is impossible. A coalition tolerates the failure of its members by using the surplus of coalition capabilities. In our research, coalition capabilities are understood as resources and services (e.g., materials, energy, power etc.). During the execution of tasks, one or more members of the coalition may fail. The paper uses the probability of the event that the coalition tolerates the failures of its members to evaluate the coalition fault tolerance. A method is proposed for determining the coalition fault tolerance in the case of multiple member failures. Complexity of the proposed method and its applicability are assessed.
Free
Scientific article
In this paper, we present the design and development of a collaborative knowledge-sharing platform with Blockchain based smart contracts (CodeBlockS) to help increase the trust and efficiency of how developers find the solution to their problems or try to learn new things. The popularity of Question-and-Answer websites such as StackOverflow, Ask, and Yahoo, as well as online course websites like as Udemy, is gradually expanding. Given this increased popularity, the quality and efficiency of user interaction must be improved such that users can try to connect with each other, ask questions about technical problems they are experiencing, or if they want to learn a topic in exchange for a fee and potentially collaborate on a project, or simply share their thoughts on a topic and improve their knowledge and network at the same time. Because these contracts will contain money, CodeBlockS has employed Ethereum Blockchain-based smart contracts to manage the data and money, as blockchain-based smart contracts are immutable and handle payments very securely. In general, social networking websites there are very few people sharing valuable knowledge and many people sharing worthless, time-consuming content that creates distraction. With the CodeBlockS system, developers find the solution to their problems or try to learn new things, and users can share their thoughts and learning on the platform. The platform also provides inbuilt smart contracts functionality using which two users can create a contract where one user will teach or solve doubt of the other user and receive fees towards service rendered.
Free
Coffee Leaf Disease Recognition Using Gist Feature
Scientific article
Coffee leaf disease recognition is important as its quality can be affected by the disease like –rust. This paper presents a coffee leaf disease recognition system with the help of gist feature. This research can help coffee producers in diagnosis of coffee plants in initial stage. Rocole coffee leaf dataset is considered in this study. Input image is pre-processed first. Resize and filtering is used as pre-processing work. Gist feature is extracted from pre-processed image. Extracted features are trained with machine learning algorithm. In testing phase, features are extracted and tested with trained ML model. Simulation is done with 10 fold cross validation. Different ML models are used and selected the best among them based on performance. SVM achieved overall 99.8% accuracy in recognizing coffee leaf disease.
Free
Scientific article
Colon cancer remains a significant global health challenge, contributing to high morbidity and mortality rates. Accurate diagnosis through histological analysis is critical for effective treatment and improved patient outcomes. In this study, we present ColoNet, a convolutional neural network (CNN)-based system designed to enhance the early detection and classification of colon adenocarcinoma using LC25000 dataset comprising 10,000 digital histopathology images. Unlike conventional CNN-based models, ColoNet integrates an optimized feature extraction strategy with deeper convolutional layers, and dropout regularization, leading to improved generalization and reduced overfitting. Additionally, the proposed model achieves faster convergence and superior classification performance compared to existing methods. The system addresses the unique challenges in distinguishing benign from malignant conditions, automating the diagnostic process and streamlining colon cancer assessments for pathologists. ColoNet was rigorously evaluated across key performance metrics, including recall, accuracy, precision, and F1-score, achieving a maximum accuracy of 96.66%. This surpasses several state-of-the-art CNN models in colon cancer classification, demonstrating its effectiveness. Its high accuracy and robust classification capabilities make it a reliable tool for identifying different colon cancer stages. By providing an efficient and automated solution for pathologists, ColoNet is expected to significantly enhance colon cancer diagnosis, supporting early detection and staging, ultimately leading to better treatment outcomes and reduced cancer-related mortality. This research underscores the importance of AI-driven systems in transforming the landscape of digital pathology and improving clinical decision-making for colon cancer.
Free
Color Histogram and DBC Co-Occurrence Matrix for Content Based Image Retrieval
Scientific article
This paper presents the integration of color histogram and DBC co-occurrence matrix for content based image retrieval. The exit DBC collect the directional edges which are calculated by applying the first-order derivatives in 0º, 45º, 90º and 135º directions. The feature vector length of DBC for a particular direction is 512 which are more for image retrieval. To avoid this problem, we collect the directional edges by excluding the center pixel and further applied the rotation invariant property. Further, we calculated the co-occurrence matrix to form the feature vector. Finally, the HSV color histogram and the DBC co-occurrence matrix are integrated to form the feature database. The retrieval results of the proposed method have been tested by conducting three experiments on Brodatz, MIT VisTex texture databases and Corel-1000 natural database. The results after being investigated show a significant improvement in terms of their evaluation measures as compared to LBP, DBC and other transform domain features.
Free
Common Fixed Point Theorem in Fuzzy Metric Spaces using weakly compatible maps
Scientific article
The aim of this paper is to prove a common fixed theorem for four mappings under weakly compatible condition in fuzzy metric space. While proving our results we utilize the idea of weakly compatible maps due to Jungck and Rhoades. Our results substantially generalize and improve a multitude of relevant common fixed point theorems of the existing literature in metric as well as fuzzy metric space.
Free
Scientific article
Knowledge processing is the prime area of information retrieval in the current era. However knowledge is subjected to the meaning of discretion in any natural language. Intelligent search in various Natural Languages is required in the huge repository of information available online. Language is the integral part for any form of communication but the language has to be meaningful. Semantics is a field of linguistics that deals with the meaning of the linguistic expressions through discovery of knowledge. In this research paper, the dictionary based approach for semantics is studied and implemented. The dictionary based proposal relies on the formalization of sentence across SVO (Subject-Verb-Object) format. Rule-based classifier helps to define the rules that are checked against the dictionary which contains sequence of Subject, Verbs and Object available in English Language. By looking at the accuracy measures, recall and precision the results obtained by the proposed approach is proven good.
Free
Comparative Advancements and Challenges of Branding Strategy in Business-to-Business Scenarios
Scientific article
Businesses have evolved with digitization and automation of manual processes through ERP solutions. These solutions fall under the category of being Business-To-Business (B2B) and help its clients to grow with a centralized solution to manage all business functions. It is easier for B2C companies to expand their reach using targeted branding strategies but takes a different approach for B2B companies to expand and grow as a Brand. It not only impacts how to reach a potential market but also impacts talent attraction and evolution into a trusted partner for potential market segments. Considering the case of an ERP company as large as SAP, various strategies are observed to target and attract potential customers as well as retain the existing clientele. This paper explores the concepts of Branding in a B2B scenario for an ERP company with a focus on SAP. It evaluates brand value, positioning, architecture, and expense comparison over the years and delves into a comparative study between SAP and one of its customers (B2C), Nestle, in terms of expenses and profits acquired. It also explores changes in Brand Strategy through Customer Advocacy, the importance of sales and marketing, and response strategies to the COVID-19 pandemic.
Free
Comparative Analysis of Explainable AI Frameworks (LIME and SHAP) in Loan Approval Systems
Scientific article
Machine learning models that lack transparency can lead to biased conclusions and decisions in automated systems in various domains. To address this issue, explainable AI (XAI) frameworks such as Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) have evolved by offering interpretable insights into machine learning model decisions. A thorough comparison of LIME and SHAP applied to a Random Forest model trained on a loan dataset resulted in an Accuracy of 85%, Precision of 84%, Recall of 97%, and an F1 score of 90%, is presented in this study. This study's primary contributions are as follows: (1) using Shapley values, which represent the contribution of each feature, to show that SHAP provides deeper and more reliable feature attributions than LIME; (2) demonstrating that LIME lacks the sophisticated interpretability of SHAP, despite offering faster and more generalizable explanations across various model types; (3) quantitatively comparing computational efficiency, where LIME displays a faster runtime of 0.1486 seconds using 9.14MB of memory compared to SHAP with a computational time of 0.3784 seconds using memory 1.2 MB. By highlighting the trade-offs between LIME and SHAP in terms of interpretability, computational complexity, and application to various computer systems, this study contributes to the field of XAI. The outcome helps stakeholders better understand and trust AI-driven loan choices, which advances the development of transparent and responsible AI systems in finance.
Free
Scientific article
Cloud-native architectures have become essential for modern application development, offering scalability, flexibility, and cost efficiency through paradigms like microservices, serverless computing, and event-driven systems. However, performance trade-offs, resource underutilization, and operational inefficiencies persist across different architectural models. This study delivers a comparative performance evaluation of four leading cloud-native architectures—Service Mesh, Event-Driven Systems, Serverless Computing, and Polyglot Persistence across AWS and GCP platforms. Using a controlled experimental setup, key performance metrics including response time, throughput, resource utilization, and operational cost (OC) were assessed under varying workloads. Serverless computing demonstrated superior cost-efficiency and dynamic scaling, though hampered by cold-start delays, while event-driven systems struck a balance between responsiveness and cost. Optimization strategies such as cold-start mitigation, adaptive auto-scaling, and hybrid storage improvements yielded significant performance gains across all architectures. The research provides critical insights for developers and system architects, offering data-driven recommendations to guide architectural choices and optimize cloud-native deployments. The study’s significance lies in its empirical approach, bridging theoretical design with real-world implementation to advance best practices in building scalable and sustainable cloud-native applications.
Free
Comparative Study of Website Page Size as Design Issue in Various Websites
Scientific article
Websites are very important means of communication in this current era of information technology. Different institutions / organizations put lots of efforts to portray complete information on beautifully designed websites. Organizations these days concerned more in providing users with all facilities online through websites, which act as an interface through which a user can get his work done without physically visiting the organization. With this the responsibility of the designer and the concerned institutions / organizations increases manifold so that the websites behavior should remain interactive and quick enough for the user to avail all facilities through websites comfortably. Speed and size of a Website are directly related with each other. Size is very important when targeting users that don't have optimal Internet connections. Author in this paper developed an online tool using .NET Framework using C# to study webpage size as Design issue in various categories of the websites like Government, Commercial, Educational, Social networking and Job portals. The automated tool developed by author function on the basis of the different standards prescribed in W3C and prescribed in analysis performed in [2]. The tool act like a parser and renders the complete code of the website and then produces result by examining the memory requirements of the component files that contribute to the total size of the website. The results produced shows that out of the five different categories of websites, it can be concluded that none of the website categories undertaken follows the recommended standards of the World Wide Web consortium showing huge violation as far as recommended page size for different websites is concerned.
Free
Comparative study of CEC’2013 problem using dual population genetic algorithm
Scientific article
Evolutionary Algorithms (EAs) are found to be effective for solving a large variety of optimization problems. In this Paper Dual Population Genetic Algorithm (DPGA) is used for solving the test functions of Congress on Evolutionary Computation 2013 (CEC’2013), by using two different crossovers. Dual Population Genetic Algorithm is found to be better in performance than traditional Genetic Algorithm. It is also able to solve the problem of premature convergence and diversity of the population in genetic algorithm. This paper proposes Dual Population Genetic Algorithm for solving the problem regarding unconstrained optimization. Dual Population Genetic Algorithm is used as meta-heuristic which is verified against 28 functions from Problem Definitions and Evaluation Criteria for the Congress on Evolutionary Computation 2013 on unconstrained set of benchmark functions using two different crossovers. The results of both the crossovers are compared with each other. The results of both the crossovers are also compared with the existing results of Standard Particle Swarm Optimization algorithm. The Experimental results showed that the algorithm found to be better for finding the solution of multimodal functions of the problem set.
Free
