International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
HMBO-LDC: A Hybrid Model Employing Reinforcement Learning with Bayesian Optimization for Long Document Classification
Full text linkWith the emergence of distributed computing platforms and cloud-big data eco-system, there has been increased growth of textual documents stored in cloud infrastructure. It is observed that most of the documents happened to be lengthy. Automatic classification of such documents is made possible with deep learning models. However, it is observed that deep learning models like CNN and its variants do have many hyper parameters that are to be optimized in order to leverage classification performance. The existing optimization methods based on random search are found to have suboptimal performance when compared with Bayesian Optimization (BO). However, BO has issues pertaining to choice of covariance function, time consumption and support for multi-core parallelism. To address these limitations, we proposed an algorithm named Enhanced Bayesian Optimization (EBO) designed to optimize hyper parameter tuning. We also proposed another algorithm known as Hybrid Model with Bayesian Optimization for Long Document Classification (HMBO-LDC). The latter invokes the former appropriately in order to improve parameter optimization of the proposed hybrid model prior to performing long document classification. HMBO-LDC is evaluated and compared against existing models such as CNN feature aggregation method, CNN with LSTM and CNN with recurrent attention model. Experimental results revealed that HMBO-LDC outperforms other methods with highest classification accuracy 98.76%
Advancements in Hardware-Enabled Cyber-Physical Systems: A Comprehensive Exploration in Electronics and Computer Science.
Full text linkThe rapid evolution of Hardware-Enabled Cyber-Physical Systems (HE-CPS) plays a pivotal role in reshaping the landscape of Electronics and Computer Science. This research delves into recent breakthroughs, aiming to elucidate the integration of state-of-the-art hardware, artificial intelligence (AI), and machine learning (ML). The backdrop underscores the growing significance of cyber-physical systems and the pressing need for advanced hardware capabilities.The research's core objective is to analyze and showcase advancements in hardware design, AI and ML integration, and the mitigation of security concerns. Methodologically, a rigorous examination of peer-reviewed literature and in-depth case studies from real-world implementations forms the foundation. These case studies encompass diverse sectors, providing genuine insights into the practical applications of HE-CPS. The findings spotlight a paradigmatic shift in hardware design, emphasizing heightened efficiency, speed, and integration capacities. The infusion of AI and ML emerges as a transformative force, enhancing adaptability and predictive capabilities. Addressing security and privacy concerns reveals tangible solutions, including robust encryption and authentication measures. Real-world case studies demonstrate successful HE-CPS implementations, illustrating tangible benefits in sectors such as healthcare and manufacturing. This research contributes substantively to the discourse on the trajectory of cyber-physical systems, offering a comprehensive overview of recent advancements
Implementation of Method for Identification of Ripening Factor of Fruit Based on Improved FCM and CNN
Full text linkIn this paper we are designing new approach for detection of artificially ripened fruit. The overall approach is based on modified fuzzy C-Means (FCM) algorithm and improved convolutional neural network (CNN). The stated method will be applicable for regular and irregular shaped fruit image which are captured in natural light. Traditional FCM algorithm is powerful and used in segmentation to segment images. Study shows that FCM can be applicable for irregular shape fruit images but shows poor results for images taken in outside environment. Our modified approach will overcome the drawback of traditional FCM and will be suitable for images which are captured in outside varying intensity light. CNN is powerful tool which can be applied on predefined dataset. We know that fruits are having different shape and size; and ripening factor varies from fruit to fruit. The dataset available of fruits are captured in uniform light; so we have to modify the parameters of existing dataset as per our requirement before applying CNN. Our modified approach will make small changes in parameters of dataset before applying it to CNN; and finally CNN will show ripening factor
Optimizing Public Transportation: A Software Design Approach for Enhanced Quality
Full text linkThis study examines methods for raising the calibre of public transport networks to increase urban mobility. To manage traffic congestion, promote sustainable transportation alternatives, and improve. overall urban livability, the research emphasizes the importance of efficient and dependable public transit. This study looks at the problems that are currently occurring. It identifies the major elements impacting the quality of public transit via a thorough assessment of the literature and empirical analysis. A mixed-methods approach is used in the study, which also includes qualitative stakeholder interviews, consumer satisfaction surveys, and quantitative analysis of passenger trends. The findings draw attention to the major problems and gaps in the current public transport systems and suggest focused development measures. This study examines methods for raising the calibre of public transport networks to increase urban mobility. To manage traffic congestion, promote sustainable transportation alternatives, and improve The study adds to the body of knowledge by revealing practical tips for raising the calibre of public transit, which may guide planning and policy-making. The findings of this study will help transportation authorities, urban planners, and legislators develop strategies for building effective, sustainable, and user-friendly public transportation systems that satisfy urban inhabitants' requirements and expectations
Incorporating Learner Emotions through Sentiment Analysis in Adaptive E-learning Systems: A Pilot Study
Full text linkThis research delves into the exciting avenue of incorporating learner emotions into adaptive E-learning systems through sentiment analysis techniques. Utilizing a pilot study with 40 undergraduate computer science students, we investigated the ability of an adaptive system to detect boredom and frustration in learner forum posts and subsequently personalize content or offer support based on these emotional states. This approach proved demonstrably successful, as learners in the experimental group who received emotion-based adaptation exhibited both increased engagement (reflected in higher time spent on tasks) and improved learning outcomes (evidenced by higher post-test scores). Furthermore, qualitative feedback revealed positive responses to the personalized interventions, indicating that learners appreciated the tailored support provided by the system. While acknowledging limitations such as the small sample size and single subject area, this study firmly establishes the promising potential of emotion-aware adaptive systems. By addressing the emotional dynamics of the learning process, such systems can pave the way for truly personalized and responsive E-learning environments that cater to individual learner needs and foster deeper engagement, positive learning experiences, and ultimately, success for all students
A Framework to Automate Requirements Specification Task
Full text linkRequirement identification and prioritisation are two principal activities of the requirement engineering process in the Software Development Life cycle. There are several approaches to prioritization of requirements identified by the stakeholders. However, there is a need for a deeper understanding of the optimal approach. Much study has been done and machine learning has proven to help automate requirement engineering tasks. A framework that identifies the types of requirements and assigns the priority to requirements does not exist. This study examines the behaviour of the different machine learning algorithms used for software requirements identification and prioritisation. Due to variations in research methodologies and datasets, the results of various studies are inherently contradictory. A framework that identifies the types of requirements and assigns the priority to requirements does not exist. This paper further discusses a framework for text preparation of requirements stated in natural language, type identification and requirements prioritisation has been proposed and implemented. After analysing the ML algorithms that are now in use, it can be concluded that it is necessary to take into account the various types of requirements when dealing with the identification and classification of requirements. A Multiple Correlation Coefficient-based Decision Tree (MCC-based DT) algorithm considers multiple features to map to a requirement and hence overcomes the limitations of the existing machine learning algorithms. The results demonstrated that the MCC-based DT algorithm has enhanced type identification performance compared to existing ML methods. The MCC-based DT algorithm is 4.42% more accurate than the Decision Tree algorithm. This study also tries to determine an optimisation algorithm that is likely to prioritise software requirements and further evaluate the performance. The sparse matrix produced for the text dataset indicates that Adam optimisation method must be modified to assign the requirement a more precise priority. To address the limitations of the Adam Algorithm, the Automated Requirement Prioritisation Technique, an innovative algorithm, is implemented in this work. Testing the ARPT on 43 projects reveals that the mean squared error is reduced to 1.34 and the error cost is reduced to 0.0001. The results indicate an 84% improvement in the prioritisation of requirements compared to the Adam algorithm
Model Predictive Control with Numerical Solution based on Contraction Mapping Method for Stabilization of Vehicle Nonlinear Dynamics
Full text linkThis paper investigates the nonlinear model predictive control problem for stabilization of unstable vehicle dynamics. Model predictive control (MPC) method is a kind of optimal feedback control method in which the control performance over a finite future is optimized. The contraction mapping algorithm is used for solving the nonlinear model predictive control problem within a short sampling period. A nonlinear tire model is employed to describe the realistic behavior of vehicle motions. The objective of this paper is to propose a nonlinear model predictive control method with a fast numerical solution algorithm called contraction mapping method for designing an automatic vehicle control system. The effectiveness of the proposed method is verified by numerical simulation
Deep Learning Methods for Tooth Detection and Classification in Various Dental Image Datasets: A Taxonomy and Future Directions
Full text linkDeep learning approaches have made significant advancements in recent years, generating considerable interest in using them for medical image analysis. In dentistry, the precision of tooth detection and classification serves as the cornerstone of dental practice as it can identify the presence of dental abnormalizes at an early stage. This paper presents an exploration of the potential of deep learning methods for tooth detection and classification across a variety of dental imaging datasets including radiographs, cone-beam computed tomography (CBCT) scans, and photograph images. Convolutional Neural Networks (CNNs) have emerged as one of the most widely used and effective deep learning methods in the field of dental disease diagnosis and medical image analysis. The study aims to conceptualize how these models can effectively learn intricate tooth features, despite having variations in tooth morphology, image quality, and imaging techniques. It highlights the increasing role of deep learning in diagnosing dental diseases and emphasizes the importance of accurate tooth classification for effective treatment planning. The study reviews existing research in deep learning-based tooth classification, discusses challenges including dataset scarcity and model interpretability, and suggests future directions
Student Performance Prediction and Classification Using Learning Analytics
Full text linkFor a productive and a good life, education is a necessity and it improves individuals' life with value and excellence. Also, education is considered a vital need for motivating self-assurance as well as providing the things are needed to partake in today's World. Throughout the years, education faced a number of challenges. Different methods of teaching and learning are suggested to increase the learning quality. In today's world, computers and portable devices are employed in every phase of daily life and many materials are available online anytime, anywhere. Technologies like Artificial Intelligence had a surprising evolution in many fields especially in educational teaching and learning processes. Higher education institutions have started to adopt the use of technology into their traditional teaching mechanisms for enhancing learning and teaching. In this paper datasets have been considered for the prediction the error ratio of student performance respectively using five machine learning algorithms. Eighteen experiments have been performed and preliminary results suggest that performances of students might be predictable and classification of these performances can be increased by applying pre-processing to the raw data before implementing machine learning algorithms
Analyzing the Performance of the Fuzzy Inference System in Decision Making
Full text linkInference systems that are fuzzy, It is common practise to make use of models such as the Mamdani and Sugeno models in order to take into consideration the presence of uncertainty and imprecision in the decision-making process. MATLAB is a well-known programming environment that provides persons who are interested in developing and implementing fuzzy inference systems with the necessary tools and strategies to accomplish their goals. In order to evaluate Diabetes Mellitus (DM), the Mamdani and Sugeno fuzzy inference systems have been developed in MATLAB. This abstract provides a brief summary of how the evaluation was carried out.The Mamdani model provides a description of uncertain data through the utilisation of fuzzy sets and is founded on language standards. Through the utilisation of the Fuzzy Logic Toolbox, users of MATLAB are able to rapidly construct and simulate Mamdani fuzzy systems. Membership functions, fuzzy rule sets, simulations, and Mamdani system optimisations can all be defined and created by users without any restrictions that are placed on them. The visualisation options that are available in MATLAB, such as the surface plot and the rules plot, help to make the behaviour of the system more understandable.For the purpose of producing inferences and predictions, the Sugeno model, also known as the Takagi-Sugeno-Kang (TSK) model, combines fuzzy principles with linear calculations. It is possible to implement Sugeno fuzzy systems by utilising the Fuzzy Logic Toolbox that is included in MATLAB. The user is able to create the linear functions that are associated to each rule after the information regarding the input-output relationships has been specified through the utilisation of linguistic variables and membership functions. Evaluation, simulation, and visualisation of the rule surfaces and output response curves of Sugeno fuzzy systems can be accomplished in MATLAB in a short amount of time. To summarise, the Mamdani and Sugeno fuzzy inference systems are capable of being constructed in an efficient manner by utilising MATLAB. There is software available for rapid system modelling, simulation, and analysis applications. The techniques of fuzzy logic that are available in MATLAB can be utilised by both professionals and academics in order to address the issue of uncertainty and imprecision in decision-making processes