42 research outputs found
A New Natural Language Processing-Based Essay Grading Algorithm
No full textThe evaluation of an English essay is one of the most significant and difficult activities that is manually carried out by knowledgeable and capable instructors and faculty members. The advancement of science and technology has made it possible to automatically evaluate an English essay by employing techniques pertaining to natural language processing. For any given English essay, the intelligent system provides a generic evaluation as well as the topic/question correlation. This evaluation is based on the NLP multiple neural network model, which was used to build the system. The evaluation of essays according to worldwide standards is the primary contribution of this innovation. Any worldwide grading system, such as the Graduate Record Examination, the International English Language Testing System, etc., is qualified to make use of the grading standard. The algorithm gives users the opportunity to test their knowledge on a range of criteria, from the most basic to the most complicated, that are included in the scoring of an English essay.
 
A QR Code-Based Real-Time Auditing System for Safe Online Data Storage
No full textUp until now, auditing systems have only had a web module; these modules are complicated and not user-friendly. Protecting sensitive data stored in the cloud requires the time-consuming and laborious procedure of encrypting all of the files. To verify a user's identification in the current system, the client must input biometric data. Next, in order to safeguard the user's identity and privacy, a signature key will be validated. One major problem with biometric data is that there are a lot of circumstances that might cause it to vary, so it can't always be matched precisely. An auditing and data storage app built for the cloud is the focus of this paper. The reference ID that the client creates is used to remotely store the financial audit data in the cloud. Using a QR code scanner, this reference ID that was generated for the client is immediately transformed into a QR code. You can access the required documents by downloading them and then opening them in a dedicated app. In the event that the client's internal storage becomes corrupted or lost, this file can be restored from the cloud
A Visual Approach for Detecting Tyre Flaws That Makes Use of The Curvelet Characteristic
No full textAutomatic flaw identification is a crucial and difficult subject in the realm of industrial quality inspection for many different types of businesses. After the tyres have been manufactured, we use the curvelet transform to do an analysis on each tyre in order to locate imperfections on the tire's outer surface. In this paradigm, deep image features can be learned, and then later used for detection, classification, and retrieval tasks using bigger coefficients in the sub-highest frequency band represented by the curvelet feature. Curvelets are a type of wavelet transform that are used to represent curvelets. We investigate image categorization challenges using deep learning with the goal of applying our findings to practical, real-world applications. The findings of the experiments demonstrate that the method that was developed is capable of accurately locating and segmenting flaws in tyre images
Protection of User Profiles in Social Networks From Unauthorized Access
No full textIn a social network, the user is responsible for providing the data in the form of a user profile, which is then used to identify the legitimate user. The information may consist of a person's name, age, gender, occupation, and address, all of which are stored by an organisation and kept up to date. When a user logs in, the user's username and password, which correspond to each other, are encrypted using a homomorphic encryption algorithm that has a one-of-a-kind key. The administrator gives the user a security code, which is subsequently entered into a system to verify the user's identity. The ElGamal algorithm is utilised in order to encrypt the conversation that takes place between two individuals. In addition to this, it is able to ensure the safety of user profiles by utilising numerous servers. Because there has been an increase in the number of data breaches, it is necessary for a company to investigate the possibility of implementing security protections for user profiles. As a result, we offer a solution for the organisation of profiles that makes use of a number of different encryption techniques
A Review on Using Machine Learning to Conduct Facial Analysis in Real Time for Real-Time Profiling
No full textThe micro facial expressions and eye blinks of the liar are analyzed by the lie detection system, which makes use of the Facial Landmark Detection System that is included in the OpenCV Tool Kit. While the suspect responds to a series of questions, the system will observe the motions of the facial muscles and the rate at which their eyes blink. The Eye-Opening Ratio is used to determine the eye-opening in each frame. An approach that makes use of human behaviors to identify deception has been proposed here. In order to assess whether a candidate is being dishonest during an interrogation and come up with a conclusion about them, the system will do face detection and an eye blink calculation. During an interrogation session, the interrogator can use this result to assist them in doing an analysis of the blink threshold value and locating the lie. In the future, developments could include thermal monitoring, which would involve collecting video of the suspect while they are answering questions during interrogation. This video would then be used in conjunction with face detection and eye blink rate to provide a more in-depth analysis of the suspect's dishonest behavior
The Application of Machine Learning to the Prediction of Heart Attack
No full textHeart illnesses are among the most significant contributors to mortality in the world in the modern era. Heart attacks are responsible for the death of one person every 33 seconds. disease of the cardiovascular system by disclosing the proportion of mortality all over the world that are caused by heart attacks. In order to forecast instances of heart disease, a supervised machine learning method is utilised. Because the incidence of heart strokes in younger people is growing at an alarming rate, we need to establish a method that can identify the warning signs of a heart attack at an early stage and stop the stroke before it occurs. Because it is impractical for the average person to often undertake expensive tests like the electrocardiogram (ECG), there is a need for a system that is convenient and, at the same time, accurate in forecasting the likelihood of developing heart disease. Therefore, our plan is to create a programme that, given basic symptoms such as age, sex, pulse rate, etc., can determine whether or not a person is at risk for developing a cardiac condition. The machine learning algorithm neural networks that are used in the suggested system are the most accurate and dependable
Utilizing Deep Learning Classification Method for the Detection of Potholes
No full textThe existence of potholes on the roadways is one of the primary factors that contribute to the occurrence of crashes involving automobiles. In order to find a solution to this issue, a number of different strategies have been explored. Among these methods are the employment of vibration-based sensors, manual reporting to authorities, and laser imaging for the reconstruction of three-dimensional space. The high cost of installation, potential danger during detection, and lack of night vision are just a few of the drawbacks of some of these systems. Researching the feasibility and accuracy of using thermal imaging to the problem of pothole detection is, hence, the goal of this effort. We have collected enough data with pictures of potholes in different weather conditions and used augmentation techniques to it. After this, a novel technique to this problem area that utilises thermal imaging the convolutional neural networks (CNN) method of deep learning was implemented. Also included is a comparison of the researcher's own convolutional neural model to pretrained models. Positive outcomes will follow from this investigation, and it will aid in directing future studies into this novel use of thermal imaging for pothole detection
Principal Component Analysis for ATM Facial Recognition Security
No full textThe Automated Teller Machine, also known as an ATM, has become the most common method by which individuals withdraw cash for their own use. The transactions that people conduct through the ATM ought to be protected, and the particulars ought to be kept private. Facial recognition will be used as the method of user authentication in this project so that Automated Teller Machine (ATM) transactions will be protected from fraudulent activities. The image of the account holder's face, as well as the faces of any beneficiaries, must be taken in well-lit conditions and then saved on a central server for face recognition to work. A camera is positioned within the automated teller machine (ATM) in such a way that it can take a picture of the person who is currently using it. After comparing the face of the cardholder to the account holder and beneficiaries, the system moves on to the PIN Validation step after recognising the cardholder's identit
