International Journal of Recent Technology and Engineering (IJRTE) ISSN: 2277-3878, Volume-8 Issue-4, November 2019 2377 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: D7139118419/2019©BEIESP DOI:10.35940/ijrte.D7139.118419 An Architecture for Automated Verification of Academic Testimonials in E-Learning Kh. Amirul Islam, Akash Nag, Sunil Karforma, Sripati Mukhopadhyay Abstract: Universities offering e-learning courses often provide their students with a hard copy of the marksheet. When that same student wants to apply for a job through the online application portal of a company, he/she must scan the marksheet and upload the scanned copy. This is a nuisance because there can be many such marksheets and not everyone has access to a scanner at home. The candidate is also required to provide the name of the University which issued the degree as well as the marks obtained, because these information cannot be extracted from the scanned marksheet image using OCR with 100% success rate due to many factors including: varying marksheet formats, presence of background watermarks, differing fonts, loss in quality during scanning, etc. The company must now manually verify each such application by matching the entered marks against the marks printed in the marksheet, which is a tedious process. In this paper, we propose an alternative approach where the data printed on the marksheet is also embedded in a digital copy of the marksheet. This digital copy, in the form of an image, can then be downloaded by the students from the University portal thereby eliminating the need for scanning. Furthermore, when this image is uploaded, the company, i.e. job provider, can easily verify the information by invoking a standard API exposed by the University (or some nodal agency), which will then extract the embedded information. This eliminates the need for any manual verification and the entire process is automated, simple, fast and hassle-free. Security features are also inherent in our approach thereby reducing any chances of fraud. Keywords : steganography, e-learning, LSB steganography, API I. INTRODUCTION Steganography may be used as efficient data security mechanism for hiding intellectual and valuable information of e-learning resources such as marksheets, admit cards, certificates, lecture materials, etc. by embedding data within another message, called the cover. Depending on the type of cover, we have image steganography, text steganography or even audio/video steganography. When the embedded data is used to identify the source of the message or in some way ensure authentication, the technique is called watermarking rather than steganography. In image steganography, one of the most popular and simple techniques is to use the least significant bit (LSB) of the color value of each pixel to embed information. Revised Manuscript Received on November 15, 2019. Correspondence Author Kh. Amirul Islam, Research Scholar, Dept. of Computer Science, The University of Burdwan, India. Email: ramiz.amirul@gmail.com Akash Nag*, Lecturer, Dept. of Computer Science, M.U.C. Women’s College, Burdwan, India.. Email: nag.akash.cs@gmail.com Sunil Karforma, Professor & Head, Dept. of Computer Science, The University of Burdwan, India. Email: sunilkarforma@yahoo.com Sripati Mukhopadhyay, Prof., Dept. of Comp. Sc. & Engg., Academy of Technology, West Bengal, India. Email: dr.sripatim@gmail.com Depending on how many bits are replaced, we have LSB-1 or LSB-2 steganographic approaches. Since the more significant bits are left unchanged, the change in the image is not noticeable by the naked eye leaving the causal error no room for suspicion. In color images, there are 3 channels, namely red, green and blue. Each of these channels usually contains 8 bits of information, and therefore, using LSB-2, we have the storage capacity of storing 6 bits per pixel. Alternatively, we may employ masking and filtering approaches, or other variations of selection mechanisms on the basic LSB scheme. Fig 1. Typical E-learning framework In Fig. 1, we present a typical architecture of an e-learning system. It is based on the client-server model, where the client can either be a learner/student, an instructor, or an administrator. The administrator is responsible for creating courses, generating and distributing marksheets, admit cards, etc. The instructors are responsible for creating and delivering lectures & lecture-materials, designing question papers for examinations, etc. The learners are the primary users of the system, who enroll in courses, watch lecture videos, sit for examinations and receive certificates on qualifying those examinations. In this paper, we focus more on the problems faced by a learner in an e-learning environment when they apply for jobs after completing their courses. The problems of the students revolve around uploading marksheets and certificates to job portals, while the more important problem of verification of these testimonials is faced by the job providers. In Section 1.A we discuss the first problem, while we discuss the latter in Section 1.B. A. Problems faced in testimonial uploading Students often get a hard copy of their marksheets after completing their online courses. When they subsequently apply for a job online, they are required to upload a scanned copy of the marksheet. Each candidate typically has more than one marksheet (e.g. starting from their secondary to post- graduation/Ph.D.) to scan, and not everyone has access to a scanner at home.