 Subjects -> INSTRUMENTS (Total: 63 journals)
|
|
|
- A Review on Smart society and innovations in education
-
Free pre-print version: Loading...
Rate this result:
What is this?
Please help us test our new pre-print finding feature by giving the pre-print link a rating.
A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Authors: Ojasvi Sharma, Pragati Porwal, Mohit Bajpai
Pages: 1 - 5
Abstract: Today, discussions over the prospects of digital technology in education are increasingly viewed as part of the larger approach to educational innovation. Despite the fact that education systems and institutions are really not slow to react, important barriers appear to exist. Preventing digital technologies from attaining their complete potential in educational institutions and teaching and learning practices. Learners can learn more effectively, efficiently, flexibly, and comfortably due to the advancement of new technology. Learners connect to digital resources via a wireless network using smart devices, immersing themselves in personalized and seamless learning. Smart education, a term used to describe learning in the digital age, has gotten a lot of press recently. The definition of smart education is discussed in this study, as well as a conceptual framework. In the world of education, smart education is a new paradigm. The goal of smart education is to increase the quality of lifelong learning for students. It promotes local, customized, and seamless learning to help learners develop their intelligence and problem-solving abilities in smart environments. Smart education will face numerous obstacles as technology advances and as a result of modern society, including educational theory, educational technology leadership, teachers' learning leadership, educational frameworks, and educational ideology. Learning takes place anywhere and at any time, and it generates a large amount of behavioral data from students. Educators face a significant problem in integrating data from many situations in smart cities and developing data-centric smart education in order to provide a smooth learning experience and customized personalized service for learners.
PubDate: 2022-04-28
Issue No: Vol. 12, No. 1 (2022)
- Classification of Images Using Support Vector Machine (SVM) Approach
-
Free pre-print version: Loading...
Rate this result:
What is this?
Please help us test our new pre-print finding feature by giving the pre-print link a rating.
A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Authors: Dr.S.Manthandi Periannasamy, Dr. S. Ravi Chand, G. Sasi
Pages: 6 - 10
Abstract: The article explains how to classify images using machine learning techniques. The Support Vector Machine with strong flexibility and the capacity to operate with a vast collection of input data was employed to complete this challenge. A software created in the MATLAB simulation environment was used to explain the model. The main difficulty that image classification is gathering a large enough training set of photos to obtain a high probability of successful recognition. The photographs in the CIFAR 100 database, which have a tiny size of 32x32 pixels and are publicly available. It has 60 000 photos organised into ten primary categories. The author's database was then utilised, which included 1000 pedestrians, autos, and road signs.
PubDate: 2022-04-28
Issue No: Vol. 12, No. 1 (2022)
- Classification Of Music Genre Using Machine Learning
-
Free pre-print version: Loading...
Rate this result:
What is this?
Please help us test our new pre-print finding feature by giving the pre-print link a rating.
A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Authors: Sravya Dukkupati, Sanka Saraswathi Datta, Vangaveti Nandini, Dr. S. Ramani
Pages: 11 - 16
Abstract: Machine Learning is a way that helps the systems to automatically learn from experience and improve the performance and predict the outcome more accurately without any requirement of being explicitly programmed. Music is one of the most significant and influential part of people’s life. Also, music is known as a universal language as it has the power to unite people from different places and cultures. This helps in the recognition of various communities and their cultures by the type of songs they compose. The aim of this work is to identify the genre of a song using a higher machine learning formula than the pre-existing ones. Genres are human-created category labels for examining or displaying music styles. With the growth in digital exhibition industry, the concept of autonomous trend division has also grown a lot in popularity in the recent years. In the case of automatic genre separation using an audio signal, this work introduces a full machine learning framework. To identify the genre, the system makes use of a Convolutional Neural Network (CNN). The CNN model is trained from end - to - end to predict the type of audio signal. The GTZAN data set is used here, which is a widely known set used for music recognition (MGR) analysis.
PubDate: 2022-04-28
DOI: 10.37591/joiti.v12i1.6240
Issue No: Vol. 12, No. 1 (2022)
- Contract and Feature Extraction of CBIR Method Using Soft Computing
Techniques in Machine Learning-
Free pre-print version: Loading...
Rate this result:
What is this?
Please help us test our new pre-print finding feature by giving the pre-print link a rating.
A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Authors: Mr. Kommu Naveen, R.M.S Parvathi
Pages: 17 - 26
Abstract: There is expansion in size of picture securing and information capacity strategies and furthermore there is expansion in data set of pictures. Prior text-based depiction and manual comment of pictures were utilized for recovery of pictures that was a tedious errand. The need of great importance is to deal with the huge assortments through proficient frameworks called content-based picture recovery frameworks. For this situation visual items in the picture like shape, plan and shade of the articles is available in the pictures are additionally considered alongside the related information with the picture. When contrasted with other regular technique for picture recovery these frameworks are more proficient and quicker. Additionally, we have presented a new framework for the separation of features using Gabor sorting that has been further refined using the Lion improvement. As a result, the SVM and decision tree techniques for cuckoo ocean and lion streamlining have been completed. The proposed strategy is tried as far as different boundaries that show further developed outcomes are accomplished involving Lion advancement when contrasted with cuckoo inquiry enhancement. Presently a days there is expansion in size of picture information base by the advancement in innovation. Advancement in picture recovery frameworks comes in presence by expansion in different capacity gadgets, high velocity web and expansion in limit. Images were physically cleared before metadata, which is the collection of labels, watchwords, and sentences used to describe them. There is a variety of information accessible in visual elements such as tone, surface, forms, and spatial data. Based on this visual input, the CBIR frameworks are used to search through a large database of images. Parts of the body have been broken down into distinct sections. An overview of CBIR and its square chart follows in the next section. The third section of this article focuses on the many obstacles or issues in previous work, as well as the recommended thinking used in this research to further enhance it. Using CLAHE, a survey of current turnout will be conducted for contrast enhancement in the fourth region. Segment 5 explains the use of ICA and Gabor channel for include extraction, while Segment 6 depicts element extraction using Cuckoo search and LION enhancer.
PubDate: 2022-05-04
Issue No: Vol. 12, No. 1 (2022)
- Numerical Simulation of Hybrid GSA Based Optimal Power Flow for Multi
Objective Optimization Strategy-
Free pre-print version: Loading...
Rate this result:
What is this?
Please help us test our new pre-print finding feature by giving the pre-print link a rating.
A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Authors: Ranveer Singh Gurjar, Bharat Bhushan Jain
Pages: 24 - 32
Abstract: Restricted nonlinear optimization in electric power systems engineering is a topic of Optimal Power Flow (OPF) that has been extensively investigated. It has been a long and remarkable history for the OPF, which was founded in the 1960s, of research and publication. Newcomers to OPF research face a challenging undertaking since there is so much information available and because OPF's popularity within the electric power systems community has prompted authors to presume a considerable deal of prior knowledge that readers unfamiliar with electric power systems may not have.. A significant area of study in the field of productive power framework control and arrangement is the execution and dependability of opf algorithms. In order to achieve a specific aim, Ideal Power Flow is directed. Specific job or multiple capacities might be specified as the aim for this capacity. When it comes to reducing the expense of fuel, we have implemented an optimum power stream in order to keep voltage and power output of the generator within the recommended point-of-limitation. Depending on the benefits and requirements, a different target may be used. Various scholars for the OPF problem have consolidated many streamlined system models, such as linear programming, non-linear programming, quadratic programming, Newton-based techniques, parametric methods, and interior point methods, in the past. Soft computing processes are now being considered for use in place of standard algorithms because of the problems they cause. In order to overcome these drawbacks, it becomes essential to develop soft computing-based optimization algorithms. There are numerous cutting-edge optimization methods like Evolutionary Programming, Genetic Algorithms, PSO Algorithm, etc., that have been proposed in writing to address the issue of over fitting functions (OPF). A particle swarm optimization algorithm has been enhanced in this proposal to reduce cost capacity while keeping imperatives within acceptable limits. The hybridization of particle swarm and gravitational search algorithms is used to make changes to particle swarm optimization. IEEE-118 bus framework is used in the suggested technique. As compared to current methods, the results demonstrate that the algorithm we developed performs better..
PubDate: 2022-02-28
Issue No: Vol. 12, No. 1 (2022)
- Using Convolutional Neural Networks (CNN) for Age and Gender Prediction
-
Free pre-print version: Loading...
Rate this result:
What is this?
Please help us test our new pre-print finding feature by giving the pre-print link a rating.
A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Authors: V.N.S. Manaswini
Pages: 27 - 32
Abstract: The network, security, and care have all become more dependent on age and gender identification. It's commonly used for children's access to age-appropriate content. To expand its reach, social media uses it to provide layered adverts and marketing. Face recognition has progressed to the point where we need to map it out further in order to achieve more usable results using various methodologies. In this study, we suggest using deep CNN to improve age and gender prediction. We show that considerable improvements may be found in a variety of applications, such as face recognition. Due to its vast applications in many facial analysis challenges, automatic age and gender prediction from face photos have received a lot of interest recently. Using the Caffe Model Architecture of Deep Learning Framework, we were able to considerably enhance age and gender recognition by learning representations using deep-convolutional neural networks (CNN). We propose a simplified convolution net design that may be employed even when learning data is scarce. In light of recent events, We demonstrate that our method greatly outperforms current state-of-the-art methods for age and gender estimation. This paper covers predicting age and gender, as well as face detection and recognition using atrained model. In internal evaluation, the Caffe framework outperforms Tensor Flow by 1 to 5 times. Following the training phase, we'll utilize the. Caffe model was taught to make predictions based on new data that had never been seen before. We'll write a Python script that uses Open CV for the project code. The trained model will detect the person's face and correctly forecast their age and gender.
PubDate: 2022-05-04
Issue No: Vol. 12, No. 1 (2022)
|
Subscription journal
