International Journal of Engineering and Advanced Technology

Journal Information
ISSN / EISSN : 2249-8958 / 2249-8958
Total articles ≅ 6,821
Archived in
SHERPA/ROMEO
Filter:

Latest articles in this journal

, M. Jayanthi Rao
International Journal of Engineering and Advanced Technology, Volume 11, pp 120-124; https://doi.org/10.35940/ijeat.a3161.1011121

Abstract:
Anomaly detection is an area of video analysis has a great importance in automated surveillance. Although it has been extensively studied, there has been little work started using CNN networks. Hence, in this thesis we presented a novel approach for learning motion features and modeling normal Spatio-temporal dynamics for anomaly detection. In our technique, we capture variations in scale of the patterns of motion in an image object by using optical flow dense estimation technique and train our auto encoder model using convolution long short term memories (ConvLSTM2D) as we are processing video frames and we predict the anomaly in real time using Euclidean distance between the generated and the ground truth frame and we achieved a real time accuracy of nearly 98% for the youtube videos which are not used for either testing or training. Error between the network’s output and the target output is used to classify a video volume as normal or abnormal. In addition to the use of reconstruction error, we also use prediction error for anomaly detection. The prediction models show comparable performance with state of the art methods. In comparison with the proposed method, performance is improved in one dataset. Moreover, running time is significantly faster.
, Arabinda Saha Partha
International Journal of Engineering and Advanced Technology, Volume 11, pp 240-243; https://doi.org/10.35940/ijeat.a3201.1011121

Abstract:
Machine learning teaches computers to think in a similar way to how humans do. An ML models work by exploring data and identifying patterns with minimal human intervention. A supervised ML model learns by mapping an input to an output based on labeled examples of input-output (X, y) pairs. Moreover, an unsupervised ML model works by discovering patterns and information that was previously undetected from unlabelled data. As an ML project is an extensively iterative process, there is always a need to change the ML code/model and datasets. However, when an ML model achieves 70-75% of accuracy, then the code or algorithm most probably works fine. Nevertheless, in many cases, e.g., medical or spam detection models, 75% accuracy is too low to deploy in production. A medical model used in susceptible tasks such as detecting certain diseases must have an accuracy label of 98-99%. Furthermore, that is a big challenge to achieve. In that scenario, we may have a good working model, so a model-centric approach may not help much achieve the desired accuracy threshold. However, improving the dataset will improve the overall performance of the model. Improving the dataset does not always require bringing more and more data into the dataset. Improving the quality of the data by establishing a reasonable baseline level of performance, labeler consistency, error analysis, and performance auditing will thoroughly improve the model's accuracy. This review paper focuses on the data-centric approach to improve the performance of a production machine learning model.
, Regi Kumar V
International Journal of Engineering and Advanced Technology, Volume 11, pp 189-194; https://doi.org/10.35940/ijeat.a3203.1011121

Abstract:
Customer satisfaction is the backbone of any business entity and supply chain optimization plays a vital role in customer satisfaction efforts. Supply chain inventory control is one of the scientific supply chain optimization methods for determining proper inventory levels at different stages or echelons of the supply chain to meet the requirements of the customers. The intention is to supply right type of material at exact time in appropriate quantities and at competitive rates. Supply chain inventory costs consist of costs to store, track and insure materials. Inventories that mishandled create substantial financial problems for a business, whether the mismanagement results in an inventory accumulation or an inventory shortage. Therefore, an examination of the right quantities to be kept in stock to meet the requirements, the strategic location, storage facilities and recordings of the goods or items should be done systematically such that the desired degree of service can be provided at competitive prices or at minimum ultimate cost. Major objective of inventory control in a multi echelon supply chain is to optimize inventory cost elements like transportation cost, carrying cost, holding cost and all other inventory related costs at all supply chain stages with an elevated service level at the end customer point. The supply chain inventory control becomes tough when the handling material is a perishable one as its deterioration rate is variable rather than constant. This article provides the study results of the deterioration rate of a perishable edible inventory at different selected environmental conditions. The focus of this article is to introduce a mathematical equation for the deterioration rate of the selected perishable inventory which is inevitable for the formulation of inventory models for its supply chain echelons.
International Journal of Engineering and Advanced Technology, Volume 11, pp 9-14; https://doi.org/10.35940/ijeat.a3119.1011121

Abstract:
In recent years researchers are intensely using machine learning and employing AI techniques in the medical field particularly in the domain of cancer. Breast cancer is one such example and many studies have proposed CAD systems and algorithms to efficiently detect cancer cells and tumors. Breast cancer is one of the dreadful cancers accounting for a large portion of deaths caused due to cancer worldwide mostly affecting women, needs early detection for proper diagnosis, and subsequent decrease in death rate. Thus, for efficient classification, we implemented different ML techniques on Wisconsin dataset [1] namely SVM, KNN, Decision Tree, Random Forest, Naive Bayes using accuracy as a performance metric, and as per observance, SVM has shown better results when compared to other algorithms. Also, we worked on Breast Histopathology Images [2] scanned at 40x which had images of IDC which is one of the most common types of breast cancers. And to work with the image dataset along with EDA we used high-end techniques like a mobile net where smote a resampling was used to handle imbalanced class distribution, CNN, SVC, InceptionResNetV2 where frameworks like Tensor Flow, Keras were loaded for supporting the environment and smoothly implement the algorithms.
Madan Mohan, , Anuranjan Mishra, Gniot Professor & Dean
International Journal of Engineering and Advanced Technology, Volume 11, pp 34-36; https://doi.org/10.35940/ijeat.a3130.1011121

Abstract:
Big Data is the way to go especially for the large companies which keep a lot of information on the systems. This paper delves into the new challenges associated with big data. It points out safety challenges on Big Data as the main issues that organizations seek to address on a day-to-day basis. These challenges include securing the trusted environments, sufficient access management, performing due diligence, combating AVI vulnerabilities, and security automation. They can be solved by maintaining strict access strategies that only allow their esteemed and responsible employees to login and also set the systems in such a way that they can detect abnormalities and allow for investigations while there is still time. The paper has addressed big data challenges as well as their solutions which are always be considered in the case of the organization as they have long-term consequences if not put into consideration.
, Mahmad Naheed, , , Aravindkumar B Harwalkar
International Journal of Engineering and Advanced Technology, Volume 11, pp 41-50; https://doi.org/10.35940/ijeat.a3131.1011121

Abstract:
In this works 3D modeling, design and safety management of high rise building using building information modeling (BIM) technology is carried out.. Initially a AutoCAD plan with all its salient features is developed following byelaws of high rise building. Then the 3D modeling and rendering of high rise building is done in the Revit architecture of the 2D plan which is imported from the AutoCAD. The analysis and design of high rise building is carried out using ETabs software. Apart from the structural design Mechanical, Electrical and Plumbing (MEP) services design is carried out using BIM technology . The layout of fire safety system is specified efficiently with use of BIM in co-ordination with MEP services. The application of BIM based design process resulted in considerable time reduction in compression with traditional design process and the holistic design of the high rise building is carried out with the compatibility of different softwares.
Ravindra Kumar, Navvis Healthcare Technical Account Manager
International Journal of Engineering and Advanced Technology, Volume 11, pp 70-72; https://doi.org/10.35940/ijeat.f3025.1011121

Abstract:
Sentimental analysis and opinion extraction are emerging fields at AI. These approaches help organizations to use the opinions, sentiments, and subjectivity of their consumers in decision-making. Sentiments, views, and opinions show the feeling of the consumers towards a given product or service. In recent years, Opinion Mining and Sentiment Analysis has become an important tool to detect the factors affecting mental health. It’s Also true that human biasness is available in giving opinions, but it can be eliminated through the use of algorithms to get better results. However, it is crucial to remember that the developers are human and might pass the biasness to the algorithms during training. The main target of this paper is to give background knowledge on opinion extraction and sentimental analysis and how factors affecting mental health can be collected. The paper aimed to use interested individuals in knowing some of the algorithms in opinions extraction and sentimental analysis. The paper also provides benefits of using sentiment analysis and some of the challenges of using the algorithms.
Ashutosh Kumar, Raghvendra Gautam
International Journal of Engineering and Advanced Technology, Volume 11, pp 60-69; https://doi.org/10.35940/ijeat.a3152.1011121

Abstract:
Objectives: To study a hybrid VTOL- Blended wing body design for its wings and elevons and perform CFD simulations with the wings. The steps for designing wing configuration and Elevon positioning involve different variables giving rise to a large number of design possibilities for a control surface. In the current study methods, have been proposed for the selection of optimized wing configuration and elevons positioning and validated with simulations model. Methods: Meta-heuristic methods like genetic algorithms are used for arriving at favorable solutions and Matlab coding is written for the initial draft of wing geometry, selected geometries are iterated in XFLR5 for stability and control, and later validated with simulations around the fluid domain. Elevons are control surfaces generally installed in tailless aircraft at the wing's trailing edge. It applies to roll and pitching force to wings as it combines the functionality of both pitching and rolling control. Design space was mathematically plotted and solved using MATLAB to decide elevons, wing configuration, and their positions.Findings: Initial selection of wing geometry, aoa, and structural design for maneuverability and stability for the enhanced aerodynamic performance of BWB UAV. In this presented paper drag coefficient of the designed BWB UAV comes out to be precisely around 0.02216 using computational modeling. Variation curve of Lift and drag coefficient with aspect ratio and angle of attack. Post-processing results of pressure forces and velocity profile on Wings accurately validate the proposed method of control surface optimization. Novelty: Designed BWB UAV has increased lift to drag ratio, reduced weight of airframe which improves performance. The Design phase is highly iterative, Through this research paper, an attempt has been made to develop a methodology for selection and investigation of control surfaces against requirements that makes BWB UAV more helpful for practical use and increasing the lift and endurance efficiency of the hybrid VTOL- Blended wing body aircraft.
Paulin Boale B., Simon Ntumba, Eugene Mbuyi M.
International Journal of Engineering and Advanced Technology, Volume 11, pp 126-131; https://doi.org/10.35940/ijeat.a3177.1011121

Abstract:
Bootstrapping is a technique that was introduced by Gentry in 2009. It is based on reencryption which allows an encryption scheme to perform an unlimited number of processing on encrypted data. It is a bottleneck in the practicability of these schemes because of multiplication operations which are costly in complexity. This complexity was reduced in TFHE by processing bootstrapping on the result of a two-bit logic gate in thirteen milliseconds using the Fast Fourier Transform. Building on this advance, an implementation of the addition of ten (10) numbers of 32-bits was performed based on the 32-bit Carry Look ahead Adder and was executed in less than 35 seconds using the configured SPQLIOS Fast Fourier transform to manipulate AVX and FMA instructions. This connector improves performance to a higher level than FFTW3 and NAYUKI.
International Journal of Engineering and Advanced Technology, Volume 11, pp 138-142; https://doi.org/10.35940/ijeat.a3181.1011121

Abstract:
A fire accident can be caused by many hazards, such as a propane tank, a defective product, a vehicle crash, or poor workplace safety. Because accidents involving fire are often unexpected and sudden, there isn’t a standard legal process for dealing with them, other than filing a negligence or workers compensation claim. This project aims to detect and monitor Fire Accident incidents well in advance and alert the surroundings to minimize the losses. This is an integration of IoT and Deep Learning Technologies, where sensors are used to collect the relevant data under the supervision of a controller unit. The controller unit collects and sends this data to a cloud database, from where the data for the Deep Learning model is fetched. This data is then used for making some insights and further predictive analytics. From the insights, many variables were found to be one of the reasons for a fire accident to take place. We considered the information about variables like Flame sensor, Temperature, Heat Index, GPS coordinates, Smoke, Type of Gases, Date, and Time for feature set generation and fed the model to a deep neural network for making future predictions. Comparing to existing conventional methods, this proposed method is different in terms of integrating deep learning with IoT. This method of approach will predict the chance of accidents priorly by classification of data.
Back to Top Top