Authors: Pavan Sharma , Aakash Muthreja , Deeksha Srivastava , Himanshu Bagle, Kritika
Certificate: View Certificate
Due to COVID-19, Online Examination has gained the most popularity over the previous year across all educational sectors. However, the institutions are having a lot of trouble with the proctoring procedures. There must be a solution found if the way we live is to become the new norm.In this project, we put up a plan to create an AI-based integrated system that can aid in avoiding exam cheating, as well as various methods and technologies that would allow the proctor to be absent from the exam at all times. Our AI- based algorithm will be able to recognize any irregularities in an exam. Our system guarantees a moral examination atmosphere free of unethical behavior. Our technology monitors the activities of students who attend school in distant locations. Our method can aid in the detection and prevention of exam-related cheating. You can conduct your entire exam without using any paper. By switching to our technology, we can significantly reduce the cost of managing and administering exams. Our solution is compatible with the conventional exam method. The emergence of the online examination system (OES) is the best example of these reforms, and the monitoring system is intended to assure the fairness and impartiality of the OES. The growth of modern educational technology has pushed changes in teaching patterns and examination patterns. The paper describes the OES and monitoring system\'s structure and operation, highlights the major issues with communication and security, and offers remedies. The network examination criteria have been met by the system. In this project, we\'ve suggested a way to stop exam cheating by creating an AI-based integrated system. We\'ve also provided several methods and tools that eliminate the need for the proctor to be present during the exam. Any unfairness in a test will be detectable by our AI-based approach. Here, we offer methods and resources that eliminate the requirement for the proctor to be present for the entire exam. Any unfairness in a test will be detectable by our AI-based approach. Here, we offer methods and resources that eliminate the requirement for the proctor to be present for the entire exam. Neural networks and machine learning are the foundation of this. Our AI- based technology will be able to recognize any unethical exam practices. Our tests showed that the suggested system is superior to the ones already in use.
A. Online Examination System Based On Proctoring And Monitoring
This online examination system is a software solution that enables any institution or industry to plan, carry out, and manage examsin a virtual setting . The Internet, an intranet, or a local area network are all viable options. The processing of results is delayed during manual examination systems, filing is a challenge, and filtering of records is challenging. Records are likely to be lost, and finding records is challenging. The system requires a great deal of time and effort to maintain, which is also very tough. One of the essential components of the online educational system is the online test. It is effective, moves quickly enough, and uses fewer substantial resources. On the basis of the web, a testing system is created. This essay explains the system's basic principles, highlights its primary functions, examines the mechanism for automatically creating test papers, and talks about the system's security .
B. Need For An Online Examination System
An electronic application is the Online Examination System. This framework will assist the college/institution in evaluating the inquiry with several alternatives and a single correct response. The college/institution can oversee the online test and provide a quick summary of the results. The question paper will be completely safe because it will be created by the examination office . Understudieshave remote access to online examination systems. It helps the inspector by minimizing the workload associated with conducting exams, reviewing answer sheets, and providing results. The machine completes every step of this work. Here, we use a client-server architecture. Access is provided by the executive to the instructorand students. Students who are familiar with the system will be able to administer exams. Exam segments come in two flavors practice exams and actual tests. Both tests may be taken by understudies. Correct responses will be highlighted in various shades. The results of the test will be generated after submission, and exams will be conducted based on the results, which will be sent to each and every understudy .
C. Energizing The Project
In the conventional framework, overseeing assessments is very monotonous labor for both the analyzer and the instructor. Up until this point, the entire exam scheduling and score evaluation process was carried out physically. The online examination framework is unquestionably an electronic framework, but. The framework aims to reduce costs associated with scheduling exams over time and totally computerize examination framework-related tasks including registration and result dissemination, leading to a high level of framework effectiveness. After encountering a large number of reference papers, we have finally come to the conclusion that we can create one examination framework that can provide simple access to establish directing exams and investigating outcomes.
D. System Proctoring And Monitoring
In the last six years, the number of internet users in India has almost doubled. As a result, many students could continue their studies, which was great for academics. Additionally, this made it possible for exams to be taken online, which introduced the idea of online proctoring at the academic level. A proctored exam enables the examiners to supervise remotely. To keep the exam's integrity, they use video, audio, and numerous anti-cheating tools. It is challenging to manually proctor an online exam when it is being taken remotely since multiple students cannot be watched at once. The concept is to develop an AI system that will monitor the student with the webcam and microphone and with that, the teacher will be able to monitor numerous students at once. During this, a teacher can physically monitor pupils using all of their senses. Additionally, the system ought to maintain a log of potential malpractices. An online proctored exam is about to start. This instrument assists educational institutions in monitoring the exam procedure and avoiding any form of cheating. Artificial intelligence-enabled technologies are being used by many online exam providers to proctor exams impartially. To ensure that the applicants do not engage in any cheating conduct, these sophisticated systems use audio and video access techniques .
E. Services For Proctoring
The following elements make up online proctoring services: Authentication
The first thing to look for is the actual candidate taking the exam, not a proxy. This can be done by automatically authenticating the candidate before to the exam using AI- based facial recognition and OCR.
II. LITERATURE REVIEW
Many of the researchers proposed many solutions and different types of algorithms to enhance the performance and reduce the cheating methods in the online examination system.
In a research paper  by the author Mr. Vidhya and Ms. Hema they proposed
The solution of artificial integrated system that will be able to detect any unfair in an examination.
According to their study they created the system that can keep a record of probable malpractices. This programmer aids educational institutions in keeping an eye on the exam procedure and preventing any form of cheating. To ensure that the candidates do not engage in any dishonest behavior, their sophisticated measures include audio and video access approaches. So, they use the different algorithm for the face detection.
They use the HCLBP algorithm which stands for Haar cascade local binary pattern algorithm its accuracy rate is not so high
Also, they use the GLCM algo for the image recognition but this algorithm is not so much accurate at the blur time of the image in research paper  by Motwani Nikhil according to their study they also Wanted to create a web-based system so that malpractices can be stopped So they created an online MCQ based system that there is no need of pen and paper and examination can be remote only by their web system. They also proposed the solution of multifaced detection so cheating can be prevented.
Limitations: They use the yolo algorithm for the multifaced detection its is correct but the accuracy rate of yolo algo is only 76.5 percent so they unable to make that kind of web system to prevent cheating.
In research paper  by Anjali Rajendra in this research paper they proposed the Solution of online exam as well as the hiring online tests so they proposed software would assist the human proctor in keeping track of the student activities. In this way, whenever a student is suspected, they would be brought to the forefront of the human proctor’s screen, and their suspicious activity fagged for later review. So, they generated the idea of Proctor-U.
The system is not fully based on artificial intelligence and need of the human Intervention. Proctor-u uses both microphone as well as camera but the accuracy is not so much high as the one who is watching if miss the cheating cause so cheating cannot be prevented.
III BACKGROUND WORK
You can conduct Online Aptitude Tests to understand thepreparation of various competitive exam. You can also conduct exams in multiple languages . The difficulty of holding English-language online exams in remote locations should be removed. It is simple to configure the system to display exam questions and answers in two languages, such as English and Hindi. Candidates preferring Hindi can select the Hindi option while the English Language can be chosen by students who want to appear for the online exam in English.
There may be a rise of the situation that internet connectivity may get lost during the examination. There are currently so many systems in which if the connectivity is lost then the system gets restarted and the candidate will not be able to give the examination where he/she left. But our system provides the facility of, if the internet connectivity stops during the examination, then after when internet gets back, the candidate can resume the examination from where he/she left.
Here we will be using a 360-degree camera for proctoring We had two options for resolving this issue: one included using a proctoring model on the client side, and the other involved using a proctoring model on the server side . The model would be supplied to individual devices that administer examinations along with the browser pages in the client-side approach. These pages would gather the necessary data, extract the necessary information from it, and send this information to the server in the background. The gadget could only collect data through the server-side mechanism, which would deliver browser pages to the device. The pages would transmit the server their raw data, and the server would then use a proctoring model to extract data from the photos. An online test is given in several locations. The smooth conduct of contact with remote candidates is crucial for resolving any problems or questions they may have about any process. This will guarantee that the exam runs without a hitch. Candidates can receive notifications from our system on the exam schedule and login information. In order to tell them about the online assessment procedure, you can contact them via email and SMS.
Solution: Online examination process security is one of the critical challenges. If you want to make sure that the right candidate is taking the exam in a safe setting without engaging in any fraud, you need follow the proper authentication and authorization procedures. To find leaks or security flaws during the exam process, the system should be able to audit the entire exam procedure. Our system has features that can help you to manage the security of the process. (i) Secure Browser ensures that the candidate cannot open any other window during the exam process. Opening other windows during online exam activity is prohibited,
A. Project Deliverables
Our Project involves two major elements. It starts by turning on a computer's camera and recording the student taking the test. This makes it possible for teachers to watch how the pupils behave and spot things like chatting to other people or looking stuff up in books that might be signs of cheating. Second, it either restricts the students' use of their computers for other purposes by disabling functions like copying, pasting, printing, and internet searches, or it either keeps track of everything they do on them, or both. "Locking down" a computer or browser refers to limiting students' access to problem-solving, reasoning ability of the individual. To know the nature of a person the suitable method is personality tests. Many multinational companies and institutes are using this methodto select the right candidates. It can be useful for the additional tools or resources. Exam recordings may also be examined by the instructor or teaching assistants .
B. Proposed Approach
The system requires a great deal of time and effort to maintain, which is also very tough. One of the essential components of an online educational system is the exam.It saves a lot of material resources, is quick enough to calculate results, and is effective . On the basis of the web, an AI-based examination system is created. We have suggested a web-based approach to track down and analyses student fraud during online exams using artificial intelligence. Face recognition technology can be used to identify a student taking an exam. If the student's face matches a saved photograph of them, the student is authenticated and permitted to take the test.
There are two key components to our project. In the beginning, it turns on a computer's camera and records the student doing the test. This makes it possible for teachers to watch how the students behave and spot actions like chatting with other people, mobile phones isdetected or looking up books that might be signs of cheating. Second, it either restricts the students' use of their computers for other purposes by preventing them from copying, pasting, printing, and conducting internet searches, or it either keeps trackof everything that students do on them or both.
There are currently a number of webcam proctoring systems in use, but they are not particularly accurate, cannot catch all conceivable malpractices, and occasionally catch malpractice for someone who never attempted it owing to a systemic defect. The main goal of this project is to create features that will upgrade and expand the current webcam proctoring system . The project's goals include developing the following features: head poses estimation, mouth opening detection, eyeball monitoring, number of people detection, mobile phone detection, and face spoofing detection. Python is used to create machine-learning models for each of these features. Each of the aforementioned features utilizes a live webcam feed acquired using OpenCV, and an output is obtained that offers details about the direction of the head and eyes, the existence of multiple people and a mobile phone, the opening of the mouth, and the occurrence of face spoofing. All of these outputs are saved as a log file, which can be used to identify any potential malpractices based on these characteristics .
A. Designing Of The System
HTML, an acronym of Hyper Text Markup Language. It is the markup language for developing Web pages. HTML is a combined form of various elements. HTML elements allows the page to render the content.
HTML has numerous applications, including web development. HTML code is used by developers to design how browsers display web page elements such as text, hyperlinks, and media files.
CSS is a stylesheet language for describing the appearance of an HTML or XML document. CSS defines how elements should appear on the screen, on paper, in speech, or in other media. CSS is designed to separate content from presentation elements such as layout, colors, and fonts. By giving more control and flexibility over how presentational characteristics are specified, this separation can increase content accessibility. It can also enable multiple website content to share formatting by defining the pertinent CSS in a separate.css file, reducing complexity and repetition in the structural content. Finally, it can enable the.css file to be cached to speed up page load times between the pages that share the file and its formatting.
SCSS, an acronym for Sassy Cascading Style Sheets. It is essentially an advanced and evolved version of the CSS language. It was created by Natalie Weizenbaum and Chris Eppstein, and it was designed by Hampton Catlin. It has more advanced features and is often referred to as Sassy CSS. SCSS assists a user in addingextra features to CSS, such as nesting, variables, and so on. Extra features make the process of writing SCSS easier & faster than he standard CSS language. The CSSfunction and code can be used by the SCSS language. The SCSS remains fully compliant with CSS syntax while also supporting the full power of SASS.
Python is an advanced programming language. It works with a wide range of programming paradigms, including structured, object-oriented, and functional programming. Python's simple, easy-to-learn syntax emphasizes readability, lowering programmed maintenance costs. Python provides support for modules and packages, which promotes programmed modularity and code reuse. For all popular platforms, the Python interpreter and sizable standard library are freely distributable in source or binary form.
G. Algorithms Used In The System
The system is automatically used to calculates the results using NLP and ANN algorithms.
V. RESULT AND DISCUSSION
The application of the examination and printing of the results immediately following the examination are two features that define the suggested programmer. We list the key benefits of the proposed program after the completion of implementation:
From the start of the examination to the end it consists of interfaces that are professional. Instructions for supporting the student during the test, such as switching between questions, knowing the time and remaining questions, and using the calculator are included . The program can save the answers of the students to each question and save the final degree.
Instructions for supporting the student during the test, such as switching between questions, knowing the time and remaining questions, and using the calculator are included . The program can save the answers of the studentsto each question and save the final degree (used in the case of student objection to the degree).
VI. FUTURE WORK
Thorough evaluation present day the literature tells that there are numerous methods to put into effect song Recommender machines. A have a look at modern-daystrategies proposed with the aid of preceding scientists and developers changed into completed. primarily based on the findings, the objectives latest our gadget had been fixed. As the electricity and blessings of modern-day AI-powered applications are trending, our challenge might beat rending generation usage. on this machine, we provide an outline of ways tune can have an effect on the consumer\'s temper and how to pick out the proper music tracks to enhance the person\'s moods. The carried-out system can detect the person\'s feelings. The emotions that the machine can come across have been glad, unhappy, angry, impartial, or amazed. After determining the user’s emotion, the proposed system provided the person with a playlist that carries music matches that detected the temper. Processing a huge data set is memory in addition to CPU in depth. this will make improvement more tough and appealing. The reason is to create this application in the most inexpensive possible way and also to create it below a standardized device. Our track advice gadget based on facial emotion reputation will lessen the efforts of modern-day customers in growing and coping with playlists.\\
 SWeaver, D., et al. (2005). Evaluation: WebCT and the studentexperience. Evaluations and Assessment Conference.  Deepankar Vishwas Kotwal, Shubham ,Rajendra Bhadke  Aishwarya Sanjay Gunjal andPuspendu Biswas,”onlineexamination system”, Volume: 0, Jan-2016.  Zhao Qiao-fang, Li Yong-fei,”Research & development of onlineexamination system”  IJCSI International Journal of ComputerScience Issues, Vol. 10, Issue 1, No 1,January 2013 ISSN (Print): 1694-0784|ISSN (Online): 1694-0814www.IJCSI.org  Xu Qiaoxia, Liu Dongsheng. Research and Design of theSecurity of Network Examination System [J]. Computer Education, vol 5, pp. 40- 42, 2010.  Song Luo, Jianbin Hu and Zhong Chen.\\\"Task based automatic examination system for sequenced test\\\". 2009 International Conference on Electronic Computer Technolgoy, 2009, pp.18- 21.  PENG JiJiang, YANG Guangzhong. \\\"Design and realization of internet Exam system based on .NET\\\". Science Technology and Engineering Vol(5),2005(20), pp. 1555- 1560.  WebBased online Secured Exam; B.Persis Urbana Ivy,A.shalini, A.Yamuna/International Journal of Engineering Research and Applications (IJERA) ISSN:2248-9622 www.ijera.com Vol. 2,Issue 1,JanFeb 2012, pp.943-944943  Alessio, H. M., Malay, N., Maurer, K., Bailer, A. J., & Rubin, B.(2018). Interaction of proctoring and student major on online testperformance. International Review of Research Open and DistributedLearning, 19(5), 166–185. https://doi.org/10.19173/irrodl.v19i5.3698  A Systematic Review on AI-based Proctoring Systems: Past, Present and Future (researchgate.net)  A Video Analytics System for Class Room Surveillance Applications Saeed Ahmed, Nirmal Krishnnan, Thanmay Ganta, Gurusamy Jeyakumar  email@example.com,firstname.lastname@example.org  Research on Abnormal Behavior Detection of Online Examination Based on ImageInformation Senbo Hu1 Xiao Jia2 , Yingliang Fu3 Dalian Maritime University College of Information Science and Technology Dalian , Liaoning Email:email@example.com,
Copyright © 2023 Pavan Sharma , Aakash Muthreja , Deeksha Srivastava , Himanshu Bagle, Kritika . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.