Blockchain-based MOOCs

A fair massive online open course system with peer grading implemented using blockchain

About Us

Nowadays, increasingly, many students learn the material of various courses through massive online open courses (MOOCs), instead of attending a university or an institute in a formal way. MOOCs have exploded in popularity among students and have generated a lot of money, especially during the COVID-19 pandemic, since universities in many regions around the world halted formal education and made use of online technologies, which included a lot of problems such as frequent disconnections in poor Internet coverage

areas of the cities and towns. As a consequence, many turned to the MOOCs available on the Internet to learn the necessary material by listening to high-definition recorded lectures, reading suggested notes, solving the suggested homework (and possibly be graded), doing the course project, connecting with fellow students using online forums, and asking questions in case needed from the instructor and/or the teaching assistants.

Learn More

Why the BLOCKMENT?

  • 01 What is our solution design?

    In Blockment, students sign up on the MOOC host website and take the course they want, as if they are taking a traditional course online. The record of each student's registration in the course will be put on a blockchain. As the student downloads or watches lectures, the records will be updated by the insertion of new log records on the blockchain. Meanwhile, homework handouts are designed and released collectively, and the students submit their answers to the website. Each homework solution will be assessed by a sample of a pool of people containing three user roles: other students taking the same course, the instructor, and the teaching assistants. After the completion of a course, the students will get a certificate in case they have eligible grades, and the certificate is pushed to the blockchain, making issuing the certificates a transparent and immutable action. The provided blockchain platform is a "complete ecosystem" of taking courses, aiming to be the next iteration of MOOC-providing websites and platforms, especially during the era of Web 3.0 decentralized apps (dApps) witnessing a significant increase in popularity and usage; the provided MOOC platform is a Web 3.0 equivalent of the current MOOC-providing apps and websites. Blockment supports registered students through the whole learning process and will reward them (issue certificates for them) for their effort in the course. In this system, enrollees start to learn the material, and throughout the progress of the course, they design new homework handouts and ask fellow students (with the help of the platform as a UX facilitator) to send their answers. After each member of the class solved the problem set given to them which was developed by other learners, they start to assess their fellow students' solutions. The assessment procedure works as follows: ⁃ Students submit their homework anonymously. ⁃ The students, the teacher, and the teaching assistants compete to assess the submitted homework. After assessing the homework, they release the papers, the details of the assessment, and the given grade to the website, which is then added as a block to the blockchain. Their knowledge score in the system and the time of their assessment determine the priority and impact of that block and the rewards they receive. The students get a small amount of score for assessing each paper. Each homework should be assessed by at least three (can vary) students. ⁃ The students act as miners, which means they check the assessment details of papers and add the verified assessments to the blockchain. Each block consists of the homework/exam questions, responses, and grading results. The miners get a score (similar to Bitcoin in the Bitcoin blockchain) for verifying the grades. Consequently, in Blockment, the winner is not who possesses better hardware, but who has a better knowledge of the course and is more proficient at assessing other solutions. To ensure this, every student gives a score to the student that assessed their paper, so that the blocks associated with students who have higher assessment scores have a higher chance to be accepted.

  • The assessment procedure works as follows:
    ⁃ Students submit their homework anonymously.
    ⁃ The students, the teacher, and the teaching assistants compete to assess the submitted homework. After assessing the homework, they release the papers, the details of the assessment, and the given grade to the website, which is then added as a block to the blockchain. Their knowledge score in the system and the time of their assessment determine the priority and impact of that block and the rewards they receive. The students get a small amount of score for assessing each paper.
    Each homework should be assessed by at least three (can vary) students.
    ⁃ The students act as miners, which means they check the assessment details of papers and add the verified assessments to the blockchain. Each block consists of the homework/exam questions, responses, and grading results. The miners get a score (similar to Bitcoin in the Bitcoin blockchain) for verifying the grades. Consequently, in Blockment, the winner is not who possesses better hardware, but who has a better knowledge of the course and is more proficient at assessing other solutions. To ensure this, every student gives a score to the student that assessed their paper, so that the blocks associated with students who have higher assessment scores have a higher chance to be accepted.

  • Our goal was to propose a new assessment method with the following properties that is applicable in both traditional learning and MOOCs:
    ⁃ Decentralize the process.
    ⁃ Ensure transparency.
    ⁃ Have a universal and clear procedure.
    ⁃ Have an extensive peer-review.
    ⁃ Achieve results faster.
    ⁃ Provide reliable infrastructure.
    An assessment system with the properties above can satisfy both students and teachers and be reliable at the same time. We defined, designed, and implemented our proposed system, Blockment, based on a blockchain platform, because it is decentralized, immutable, and it can guarantee transparency and fairness, unlike the other mentioned potential solutions. Consequently, Blockment helps to provide a fair and faster grading process by using a decentralized system core.

  • We expect Blockment to improve assessment quality, increase grading speed and reduce bias in academic environments when grading. In other words, our blockchain-based MOOC and peer grading system should be able to help students achieve fairer grades in less time. We back our claims by designing and launching experiments on students in a classroom environment. We designed a two-step experiment scheme. In the first phase, we generally evaluated whether our newly proposed assessment method leads to promising results by conducting experiments on a few homework handouts in certain classes, and in the second phase, after making sure the results from the first phase has been promising, we launched a bigger study on a class for a complete educational semester to be able to both evaluate longer-term effects of using a blockchain-based peer assessment system and see if positive results seen in the first experiment stays true for the duration of the whole educational semester. The results from the experiments indicated that Blockment was able to increase grading speed and decrease the assessment bias significantly (p-value < 0.05), which indicates Blockment can be used as a MOOC platform in a real-world setting to provide peer grading support.

  • As far as we know, the idea of our proposed blockchain is new and unprecedented, so we consider some policies before implementing it to provide a better user experience. For example, the system should have a good abstraction (e.g. hiding credit / scoring system implemented as part of the decentralized network) since users may not have a blockchain technology background. Who revises the grades should not be the person who graded in the first place. Each paper might be graded multiple times to find and correct errors. Details of grading should be included in the block. Students should be able to consult the instructor when needed. To prevent the destructive effects of fake users, users are rated and validated by the reputation scores. To handle biased scores, other users validate previous assessments. To address the lack of checkers and correctors, users are motivated by scores given as rewards. Other problems in common with common blockchain networks (e.g. Bitcoin) will work out the same way it did for them.

  • Some of the main concerns in the assessment of massive online open courses are the fairness of the grades and the time it takes to grade all students' answers, especially because MOOCs tend to include more than hundreds of students, while homework in certain theoretical courses can't be auto-graded by a computer program and needs the intervention of human assessors which is simply not available in the scale necessary to support all users registered in the MOOC. The validity, authority, and privacy of the certificates earned from passing MOOCs have also been challenging. In courses with fixed homework handouts and fixed solutions, some people who have passed the course previously may share solutions on social media, making it straightforward for other students to pass the exams without learning a lot and earn a certificate they didn't deserve, resulting in a kind of degree fraud. Consequently, many institutes or companies doubt the validity of these certificates and try to reassess the candidates by performing time-consuming interviews, which decreases the value of enrollment in a MOOC in the first place. MOOCs provide a lot of flexibility in students' schedules, are usually up-to-date, and are accessible to a much wider population than the traditional courses, but suffer from insufficient validity and transparency of the certificate, and assessing problems regarding hundreds of students in each semester. We claim that by using Blockment, the quality of the course, both in MOOCs and the traditional method can be improved by addressing these issues.

  • Our objective was to provide a brand-new evaluation technique with the following characteristics that would work for both conventional education and MOOCs:
    ⁃ Distribute the procedure.
    ⁃ Ensure openness.
    ⁃ Have an easy-to-follow, universal method.
    ⁃ Have a thorough peer review.
    ⁃ Get results more quickly.
    ⁃ Offer a solid infrastructure.
    A dependable and fair evaluation system that satisfies both instructors and students simultaneously has the abovementioned characteristics. In contrast to the other indicated potential solutions, we created, built, and constructed our suggested system, Blockment, based on a blockchain platform. This is because it is decentralized, irreversible, and can provide transparency and justice. As a result, Blockment uses a decentralized system core to deliver a fair and quick grading process. Several universities and institutes are now using blockchain technology in their educational systems, and most do so to assist academic degree administration and the digital transformation of certification procedures. Paper certificates' fingerprints (unique hashes) are written on the blockchain for educational systems, which reflects paper certificates as digital certificates. The blockchain also stores information about the certification authorities' and certifiers' identities. The management of certification authorities and certifiers, as well as the monitoring or revocation of certifications, are all supported by intelligent contracts. MIT Media Lab Learning and the University of Nicosia are two universities that focus on using blockchain to issue digital certificates. Compared to the prior instances, we think blockchain technology has far greater potential for use in education and can address various problems in conventional educational institutions. To provide a fair academic evaluation environment and solve other issues, such as those listed above, and add additional useful features, we suggest placing the entire process of taking a massive online open course (MOOC) on a blockchain system called Blockment.

  • During the initial stages of its appearance, blockchain technology was not able to draw much attention. However, as Bitcoin continues to run safely and steadily over the years, society has since become aware that the enormous potential applications of the underlying technology of this invention are not only in cryptocurrency but also in many other areas, such as education. Blockchain technology has become a hot topic for more and more institutions, enterprises, and researchers all around the world. Nowadays, some universities and institutes have applied blockchain technology in their educational systems, and most use it to support academic degree management and digital transformation of certification processes. The blockchain for educational platforms represents paper certificates as digital certificates, and their fingerprints (unique hashes) are written on the blockchain. In addition, the identities of certification authorities and certifiers are also stored in the blockchain. Finally, intelligent contracts support the management of certification authorities and certifiers and the monitoring or revocation of certificates. The University of Nicosia and the MIT Media Lab Learning are examples of institutions that work on using blockchain in issuing digital certificates. In addition, Alipour et al. (2022) have explored decentralized peer grading schemes but have not implemented them in a MOOC platform, as the current work does. We believe that blockchain technology has much more potential in education than the examples mentioned earlier, and it can solve many issues in commonplace educational systems. We propose putting the end-to-end procedure of taking a massive online open course (MOOC) up to earning the certificate on a blockchain system called Blockment to provide a fair environment for academic assessment while also eliminating other problems, as mentioned above, and introducing new helpful features.

 

The averages of scores given by the students.

Fairness of student's own grade in our method 84.9%
Fairness of student's own grade in traditional method 78.9%
Overall fairness of all grades in our method 81.3%
Overall fairness of all grades in traditional method 73.8%
Grading speed in our method 77.6%
Grading speed in traditional method 54.4%

Team

Iman Mohammadi

Computer Engineering student at Sharif University of Technology

Soroush Jahanzad

Computer Engineering student at Sharif University of Technology

Sina Elahimanesh

Computer Engineering student at Sharif University of Technology

Parimehr Morassafar

Computer Engineering student at Sharif University of Technology

Parsa Neshaei

Computer Engineering student at Sharif University of Technology