The ever-expanding scope of Big Data analytics in Education 

The attributes that lend beauty to a conventional classroom are many: passionate lectures, heated debates, camaraderie among students, face to face interactions, and so on. However, the model has its own shortcomings: for one, it is one-to-few. One teacher shares his or her own knowledge and experience obviously to a small number of students who are present in the classroom. Even then, typically a few among the students who match the wavelength of the teacher, the pace of the teaching, and prerequisite for comprehension get benefitted. The rest may fall through the cracks.

In this context, the advent of Big Data analytics proves to be a harbinger of change. Its applications in education are as diverse and as exciting as they are in any other domain. To a student, Big Data analytics could mean a personalized journey of learning, beginning from his or her current level of understanding of any concept. It can ‘handhold’ the student throughout the learning process. When a student is stuck, say in factors in mathematics, the system could choose the subject of quadratic equations and line up suitable explanations.

To a teacher, it means coming up with a test paper instantly. The teacher just has to define rules or patterns of the test, the system can pick and order questions. With real-time analysis of demographics, attendance, scoring patterns, the system can give live feedback both to teachers and students. It can alert teachers when a student’s participation decreases. All these make it easy to predict drop-outs.

And to the administrators of an educational institution, it could mean gaining visibility of key drivers of enrollment or finding trends, patterns, and clusters of students spanning across geographies. They can even do competition analysis and the impact of competition on their own growth.

The potential of Big Data analytics is endless. In future there can be more online and federated models that would provide a faster and localized experience to the user. We may soon have systems that can write course content that may or may not require some finetuning from a human expert. In short, the entire value chain of education can witness productivity gains from Big Data analytics.

MOOCs: the enabler

But, for all the talk of mindblowing possibilities, Big Data analytics could have remained a non-starter in a traditional classroom scenario. The core fundamentals of Big Data are defined by 5 Vs: volume, velocity, variety, veracity and value. Hence, in the absence of the required scale, employing Big Data analytics will not be effective. for A simple ‘online analytical processing’ (OLAP) system would have been better suited.

However, the emergence of Massive Open Online Courses (MOOCs) came as a game changer. In recent years, there has been an explosion of MOOCs learning platforms and content that serve millions of students each day. Every institution worth its salt has opened a shop in the great global MOOCs bazaar. These platforms generate terabytes of data each hour with billions of data points, enough to feed the systems of Big Data analytics.

The challenges

Effective use of Big Data analytics calls for a clear definition and monitoring of data handling and transformation processes. Otherwise, ‘data lakes’ become a dumping ground. It is worth remembering that the quality of output is determined only by the quality of the input. Garbage-in, garbage-out.

It is also important to ensure that data the system handles is up-to-date. While it is good to retain all the historic data, the value diminishes over time and the computational expenses far exceed the ROI.

Unlocking the possibilities

Rare insights into relevant areas are the stuff success is made of. The key appeal of Big Data analytics is its promise of delivering rare insights on demand and on scale, by determining correlation of various events that would appear rather unrelated to the human mind.

As with any entity in any sector, educational institutions need insights into values they offer, their customers, and all the transactions they carry out with their stakeholders. Hence, investment in Big Data analytics is inevitable if they want to thrive in the markets of the future.

While the cases of Big Data applications are endless, the journey begins right from the first interaction of the user with the system. The institutions must have a well thought out plan on what data to collect, when, and at what frequency, and set the system to perform specific kinds of analytical operations. Otherwise, without a strategy, Big Data analytics will be just an expensive, vanity exercise of enormous data collection.

In contrast, institutions who are well prepared and well equipped for the roll out of Big Data analytics can have the opportunity to redefine the scope of education, and its reach beyond what is considered possible today.

The article is authored by Varun Malhotra, VP- Engineering, Aakash Educational Services