Cloud-Based Big Data Analytics in Smart Educational System

Cloud-Based Big Data Analytics in Smart Educational System

Newlin Rajkumar Manokaran (Anna University – Coimbatore, India), Venkatesa Kumar Varathan (Anna University – Coimbatore, India) and Shalinie Deepak (United Institute of Technology, India)
Copyright: © 2018 |Pages: 11
DOI: 10.4018/978-1-5225-3015-2.ch011
OnDemand PDF Download:
List Price: $37.50


In this modern Digital era, Technology is a key player in transforming the educational pedagogy for the benefit of students and society at large. Technology in the classroom allows the teacher to deliver more personalized learning to the student with better interaction through the internet. Humongous amount of digital data collected day by day increases has led to the use of big data. It helps to correlate the performance and learning pattern of individual students by analysing large amount of stored activity of the students, offering worthwhile feedback etc. The use of big data analytics in a cloud environment helps in providing an instant infrastructure with low cost, accessibility, usability etc. This paper presents an innovative means towards providing a smarter educational system in schools. It improves individual efficiency by providing a way to monitor the progress of individual student by maintaining a detailed profile. This framework has been established in a cloud environment which is an online learning system where the usage pattern of individual students are collected.
Chapter Preview


Big Data

Analogously, the information and data have gone digital and is being hovering around. The amount of information that is rendezvous and stored seemed to be increasing triple folds every day. Although web makes the data apparent to interact, big data is more than just communicating. It has the capability to put forth many aspects of the world into the data that has never been calibrated before. This process is termed as “datafication”. These volumes of data undergoes disparate transitions by collecting large amount of data rather than smaller samples, then setting predilection for a highly curated data, and lastly discovering various correlation among those data. This upsurge and explosion of data led to the rise of big data. Big data deals with a large and superfluous amount of data. The data is too big, too fast and does not fit the framework in the traditional data structure. Few applications of big data are stock market sentiment analysis, Picasa photo storing, recommendation systems, customer based analytics, machine generated data, booking systems, social networking for entertainment, medical diagnostics. Predominantly, big data is characterised by three ‘V’s that describe them. They are volume, velocity and variety.

The volume portrays the amount of data that are processed. It consumes colossal amount of data. These data are so mammoth that the conventional relational databases are hard to cope up with. There is growth in data deliberately that the database should be scaling to keep up with the growth. It is noted that big data handles more than 1 million transactions and can store a plethora of data to approximately 312 terabytes for a single unique transaction.

The velocity of the data explains the momentum in which they are processed. Various industries have to deal with the fast moving data flow into their system. Since the data are mostly processed in real time, the capability to process those data at high speed is required. It does not explain just the speed or the velocity of the incoming data, it embodies the surge of data into bulk storage for a later batch processing.

Variety depicts the disparate data that vary from structured data like numeric data in traditional databases to unstructured data that include text documents, email, video, audio, stack ticker data and financial transactions. Big data is the place where all types of data are being processed. It requires new technologies and techniques to capture, store, and analyze it. Figure 1 shows the amount of structured and unstructured data. Roughly, the unstructured data comprises of almost 80% of the databases while the structured data comprises of only 20%.

Figure 1.

Structured and unstructured data

In conjunction with these 3 V’s, big data includes veracity and value.

The big data analytics here is done providing a sophisticated way of accessing the students by predicting the behaviour of the student. With those predictions as a feedback, personalized and appropriate learning materials are issued as well as help the teachers to intervene if necessary. The database used here is HBase and the data analytics are performed in a hadoop environment. This educational system provides a holistic approach by gradually moving towards teacher centric to the student centric environment.


Cloud Computing

Cloud computing is the one that reckon on sharing computing resources rather than having local servers or personal devices to handle applications. There are few traits that makes the cloud computing desirable. First of all it can be easily managed; secondly it can be got on demand which may be public, private or hybrid. The services offered on a cloud are software as a service, infrastructure as a service and platform as a service.

These above discussed technologies are expeditiously emerging to deliver powerful results and benefits for most of the forward thinking companies. By converging the big data analytics operations into cloud environment, firms can increase the efficiency and productivity, while saving time and effort. This cloud-based approach to big data analytics serves to be more beneficial as it helps reduce cost, to take fast and better decision making and to create prominent new products and services.

Complete Chapter List

Search this Book: