International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
A Survey on Artificial Intelligence based Methods for Locating Hubs in Transport Networks
Get PDFThe location of hubs in transport networks constitutes one of the key elements affecting the organization of freight and passengers’ transport logistics activities. This work offers a good literature review of implementing artificial intelligence (AI) methods in identifying hubs of such networks. In this research study, the author examines different categories of AI techniques such as machine learning techniques, innovative neural structures, and optimization techniques to understand how those technologies could be useful for the improvement of hub location techniques. The given survey offers the comparison of various AI techniques and encourages potential applicants for showing on real transport circumstances where different sorts of AI let in the construction of good consequences. Unlike most prior papers in the context of AI-based hub location, this research contributes not only a literature review of theories but also discussion about data needs, algorithms, and interfaces with the current transport systems. Based on the analysis of the results of the latest studies and the definition of new trends related to the use of AI, this survey will also be useful for researchers and practitioners who are interested in the application of AI in the effective management of transport networks. From the findings of this study, the following lessons are anticipated to support advancement in stronger, cheaper and more convenient transport solutions that will improve accessibility and increase economic recovery
A Detailed Review on Fault Diagnosis of Electronic Systems Using Intelligent Techniques
Get PDFThis work look at different smart ways to find problems in electronic things, which getting more important as electronic stuff get more complex and need to work better. We talk about five main ways: Rule-Based, Model-Based, Case-Based, Fuzzy Logic and Neural Networks, and Hybrid Approaches. Each way have good and bad points. Rule-Based use expert knowledge but hard to keep up. Model-Based try to copy how things work but often too slow for big systems. Case-Based learn from old problems but need lots of examples. Fuzzy Logic and Neural Networks good with unclear stuff but sometimes hard to understand. Hybrid Approaches mix these ways to get the best parts of each. We look at how these ways work, where they used, and what problems they have. We also talk about what might happen with these ways in the future. Smart ways to find problems help electronic things work better and cost less to fix. They used more and more in important areas like flying, health care, and big machines. These ways can look at lots of information fast and find problems quick, which really important for keeping things safe and working. The history of using smart ways to find problems in electronic things go back many years. It start with simple computer thinking in the 1980s and 1990s. Then it get better with new math ideas in the 2000s. Now, with big computer power and lots of data, machine learning getting really good at finding problems. As electronic things keep getting more complex, these smart ways to find problems will probably get even more important. The big goal is to make electronic things that can find and fix their own problems, so they work better and need less fixing by people
From Raw Data to Actionable Insights: Leveraging LLMs for Automation
Get PDFThis paper explores the transformative role of Large Language Models (LLMs) in automating the data processing lifecycle, from ingestion to insights generation. LLMs streamline data handling by automating ingestion, transformation, and modeling processes, offering efficient, reliable, and timely insights critical for sectors such as healthcare, finance, and telecommunications. This study details the technical architecture of LLM-driven data workflows, addresses challenges in integrating diverse data sources, and emphasizes the necessity of governance frameworks to mitigate ethical concerns about data privacy and bias.
However, the integration of LLMs also presents specific challenges, such as handling unstructured data, ensuring data quality, and managing computational costs. Through case studies across multiple industries, this study illustrates the benefits and limitations of LLMs, highlighting both technical and ethical considerations for deploying these tools at scale. Case studies include a healthcare provider improving patient diagnosis accuracy, a financial institution enhancing fraud detection, and a telecommunications company optimizing network performance. Each case study employed a methodology involving data preprocessing, LLM training, and evaluation metrics to measure performance improvements. The quantitative results show the significant impact of LLMs on the data workflow
Customized Privacy Settings: Empowering User Preferences in Social Media Permissions
Get PDFIn the rapidly evolving the digital landscape of social media, user consent and data privacy have emerged as critical facets of social media interaction. This study addresses the complexity of privacy management within social media applications by probing into user preferences for permission requests. With the objective of streamlining the privacy settings process, the research seeks to understand patterns in user consent and to develop an approach that enhances user engagement without compromising data protection.
Utilizing hierarchical clustering and machine learning techniques on a dataset comprising various social media permissions, we identified four principal clusters. These clusters signify distinct user patterns in granting permissions, reflecting diverse attitudes towards privacy that challenge the conventional one-size-fits-all privacy framework.
Our methodology involved condensing the vast array of permissions into a manageable set. By refining the permissions queried from 46 to 10, our predictive model maintained high accuracy while substantially improving the likelihood of users completing the privacy settings process. This reduction led to a more personalized and less cumbersome user experience.
The study's key findings reveal significant variability in user concerns, ranging from pronounced apprehension to relative indifference regarding permissions. These findings hold substantial implications for privacy management, suggesting a need for customizable privacy settings that align with individual user preferences.
The significance of our research lies in its potential to guide app developers and policymakers in enhancing user trust and satisfaction. By aligning privacy practices with user expectations, this study contributes to the broader dialogue on user-centric privacy approaches in social media and presents a pathway to fostering more secure and personalized digital environments
AI Vs. Traditional IDS: Comparative Analysis of Real-World Detection Capabilities
Get PDFArtificial intelligence has transformed cybersecurity, particularly intrusion detection systems. Conduct a comparative analysis of AI-driven and conventional Intrusion Detection Systems (IDS). Enhance the identification of intricate cyber intrusions. AI-driven Intrusion Detection Systems enhance detection accuracy, reduce false positives, and optimize response times with Machine Learning and Deep Learning.
Intrusion Detection Systems (IDS) identify intrusion trends in historical data utilizing decision trees, support vector machines (SVMs), and ensemble methods. These algorithms enhance IDS detection models by adapting to evolving data to identify emerging threats. RNNs and CNNs more effectively manage high-dimensional network and system logs.
The essay thoroughly contrasts AI-driven Intrusion Detection Systems with signature-based and anomaly-based Intrusion Detection Systems. Conventional Intrusion Detection Systems employ signatures and algorithms to identify threats. They identify attack patterns but not intricate assaults. AI-driven Intrusion Detection Systems identify novel attack vectors inside extensive datasets via sophisticated algorithms.
Assault scenarios and datasets evaluate actual performance in real-world contexts. AI-driven Intrusion Detection Systems identify DDoS attacks, malware, and insider risks. The efficacy of AI-powered systems is contingent upon detection accuracy, false positive rates, and response times.
Integrating AI with Intrusion Detection Systems is challenging. The essay addresses the training and validation of AI models, encompassing overfitting and the necessity of extensive, representative datasets. AI-driven Intrusion Detection Systems on extensive networks may require substantial computational resources and infrastructure, so scalability is considered. Trust and accountability in cybersecurity necessitate the interpretability and openness of AI models.
IDS monitors artificial intelligence technologies. Research indicates that a hybrid AI-based Intrusion Detection System may be more effective. It also examines how XAI enhances the interpretability of AI-powered IDS and facilitates corporate adoption.
An extensive study on AI-powered Intrusion Detection Systems reveals performance metrics, implementation obstacles, and potential opportunities. The results indicate that the transformation of cybersecurity by AI necessitates additional investigation. AI-enhanced Intrusion Detection Systems mitigate the limitations of traditional IDS to enhance cybersecurity
Anemia Detection using a Deep Learning Algorithm by Palm Images
Get PDFOur aim is to detect anemia through a comparative analysis of three convolutional neural network (CNN) models, namely EfficientNet B3, DenseNet121, and CNN AllNet. A collection of 3,000 microscopic palm pictures, including 1,500 anaemic and 1,500 non-anemic samples, was used to train and test the algorithms. The dataset was preprocessed to balance the classes, augment the images, and normalize the pixel values. The models were trained using transfer learning on the ImageNet dataset and fine-tuned on the anemia dataset. The performance of the models was evaluated based on accuracy, precision, recall, and F1-score. The results showed that CNN ALLNET achieved the highest accuracy of 96.8%, followed by DenseNet121 with 94.4%, and EfficientNet B3 with 91.2%. The precision, recall, and F1-score also followed a similar trend. The study concludes that CNN ALLNET is the optimal model for anemia detection due to its high accuracy and overall better performance when compared with the different models. The findings of this research could provide a basis for further studies on anemia detection using CNN models, ultimately improving the accuracy and efficiency of anemia diagnosis and treatment
Reinforcement of the Bank Loan Model using the Feature Selection Method of Machine Learning
Get PDFDoes feature selection and machine learning (ML) guarantee the effectiveness of the bank credit system model? This article aims to analyze this problem. In fact, in finance, expert-based credit risk models still dominate. In this study, we establish a new benchmark using consumer data and present machine learning methods. A risk prediction that is as accurate as possible is an important requirements for credit scoring models. In addition, regulators expect that the models should to be auditable and transparent. As a result, the superior predictive power of contemporary machine learning algorithms cannot be fully utilized in credit scoring because very simple predictive models, such as several ML classifiers, are still widely used. As a result, significant potential is missed, increasing reserves or the number of credit defaults. A framework for comparing scores before and after feature selection machine learning models that are transparent, auditable, and explainable is presented in this article, as well as the various dimensions that need to be taken into consideration in order to make credit scoring models understandable. In accordance with this framework, we give an overview of the models which demonstrate how it can be used in credit scoring, and compare the results to scorecards' interpretability. The model presented demonstrates that machine learning techniques can maintain their ability to enhance predictive power while still maintaining a comparable level of interpretability
Resource Allocation Challenges and Strategies for RF-Energy Harvesting Networks Supporting QoS
Get PDFThis paper specifically addresses the resource allocation challenges encountered in wireless sensor networks that incorporate RF energy harvesting capabilities, commonly referred to as RF-energy harvesting networks (RF-EHNs). RF energy harvesting and transmission techniques bring substantial advantages for applications requiring Quality of Service (QoS) support, as they enable proactive replenishment of wireless devices. We commence by providing an overview of RF-EHNs, followed by an in-depth examination of the resource allocation challenges associated with this technology. In addition, we present a case study that focuses on the design of an efficient operating strategy for RF-EHN receivers. Our investigation highlights the critical aspects of service differentiation and QoS support, which have received limited attention in previous research. Besides, we explore previously unexplored areas within these domains
A Word Embeddings based Approach for Author Profiling: Gender and Age Prediction
Get PDFAuthor Profiling (AP) is a method of identifying the demographic profiles such as age, gender, location, native language and personality traits of an author by processing their written texts. The AP techniques are used in multiple applications such as literary research, marketing, forensics and security. The researchers identified various differences in the authors writing styles by analysing various datasets. The differences in writing styles are represented as stylistic features. The researchers extracted several style based features like structural, content, word, character, syntactic, readability and semantic features to recognize the profiles of the authors. Traditionally, the researchers extracted various feature combinations for differentiating the profiles of authors. Several existing works are used Machine Learning (ML) methods for predicting the author characteristics of a new author. The existing works achieved good accuracies for predicting the author characteristics by considering the both stylistic features and ML algorithms combination. Recently, in advent of Deep Learning (DL) techniques the researchers are proposed approaches to author profiling by using these techniques. Few researchers identified that the deep learning techniques performance is good for author profiles prediction than the results of style based features. In this work, a word embeddings based approach is proposed for gender and age prediction. In this approach, the experiment conducted with different word embedding models such as Word2Vec, GloVe, FastText and BERT for generating word vectors for words. The documents are converted as vectors by using the document representation technique which uses the word embeddings of words. The document vectors are transferred to three different ML algorithms such as Extreme Gradient Boosting (XGBoost), Random Forest (RF) and Logistic Regression (LR) for generating the trained model. This model is used for predicating the accuracy of age and gender prediction. The XGBoost classifier with word embeddings of BERT achieved good accuracies for age and gender prediction than other word embeddings and ML algorithms. The experiment implemented on PAN 2014 competition Reviews dataset for age and gender prediction. The proposed approach attained best accuracies for predicting age and gender than the performances of various existing approaches proposed for AP
An Automated System for Depression Detection Based on Facial and Vocal Features
Get PDFDiagnosing depression is a challenge due to the subjective nature of traditional tools like questionnaires and interviews. Researchers are exploring alternative methods for detecting depression, such as using facial and vocal features. This study investigated the potential of facial and vocal features for depression detection using two datasets: images of facial expressions with emotion labels, and a vocal expression dataset with positive and negative words. Four deep-learning models were evaluated for depression detection from facial expressions, and two traditional machine-learning models were trained for sentiment analysis on the vocal expression dataset. The CNN model performed best for facial expression analysis, while the Naive Bayes model performed best for vocal expression analysis. The models were integrated into a web application for depression analysis, allowing users to upload a video and receive an analysis of their facial and vocal expressions for signs of depression. This study demonstrates the potential of using facial and vocal features for depression detection and provides insight into the performance of different machine learning algorithms for this task. The web application has the potential to be a useful tool for individuals monitoring their mental health and may support mental health professionals in their clinical assessments of depression