A Detailed Study on Classification Algorithms in Big Data

A Detailed Study on Classification Algorithms in Big Data

Saranya N. (Sri Eshwar College of Engineering, Coimbatore, India) and Saravana Selvam (PSR Engineering College, India)
Copyright: © 2020 |Pages: 17
DOI: 10.4018/978-1-5225-9750-6.ch002

Abstract

After an era of managing data collection difficulties, these days the issue has turned into the problem of how to process these vast amounts of information. Scientists, as well as researchers, think that today, probably the most essential topic in computing science is Big Data. Big Data is used to clarify the huge volume of data that could exist in any structure. This makes it difficult for standard controlling approaches for mining the best possible data through such large data sets. Classification in Big Data is a procedure of summing up data sets dependent on various examples. There are distinctive classification frameworks which help us to classify data collections. A few methods that discussed in the chapter are Multi-Layer Perception Linear Regression, C4.5, CART, J48, SVM, ID3, Random Forest, and KNN. The target of this chapter is to provide a comprehensive evaluation of classification methods that are in effect commonly utilized.
Chapter Preview
Top

Introduction

Big data is which grows each year in a system. Invention in research, as well as scientific factors, has impacted the size of data that grows every day with the goal to enhance profitable activities. On the other hand, information accessibility and exploration of thoughts have resulted in the delivery of a completely progressive modification of the internet, which has been utilized in the most beneficial ways. As a result, billions of bytes of information are created every day hence throwing much information at regular intervals. Since newly provided data night is unstructured, structured or possibly confused; we need classification techniques in case of unstructured or complex data. A classification is an approach that allots data in a class to focus on Categories or groups (Owais & Hussein, 2016). The objective of grouping would be to precisely estimate the prospective class of every single file in the data.

Diverse Classification strategies make use of distinctive methods intended for finding the relationships among the list of data units. These types of relationships are generally compacted in a model, which may then have the ability to connect to an alternate dataset when class assignments are covered up. Classification models are broken down through differentiating the expected valuations as well as known values in a wide scope of test data (Revathy & Lawrance, 2017). The target data to get a classification model is normally isolated into several sets of data: one for understanding the genuine model; another for testing the actual model.

The overall classification technique is shown in Figure 1 below. The purpose of the learning stage is to assemble the classification model and testing stage. In the learning stage, a classification strategy is produced to portray a redid set of data classes. In both learning and training phase, the classification approach develops the classifier using a method of comprehension from the training dataset and the related class names.

Figure 1.

Classification process

978-1-5225-9750-6.ch002.f01

Organization of this Paper

This paper is composed of classification processes, issues in classification and different classification algorithms used to examine the different classification algorithms. Finally, the last section concludes the paper.

Top

Machine Learning And Classification Process

Machine learning is a quickly developing area, which shows how computers able users to learn or enhance the performance to utilize data. Computer programs naturally figure out how we can distinguish patterns and settle on accurate decisions with respect to data (Wu, Zhu, Wu & Ding, 2013). We represent great issues in machine learning, which are profoundly identified with data mining and big data. There are three primary sorts of machine learning ideas as shown in Figure 2. These ideas are described as following:

Supervised Learning

Supervised learning comprises of all the information for algorithms to figure out how output can be predicted from training dataset. This idea is utilized in the classification of problems (Revathy & Lawrance, 2017) e.g. Naive Bayes, Random Forest, Support Vector Machine (SVM).

Unsupervised Learning

Clustering-based techniques use unsupervised learning whereby information remains unlabeled. The output structure is figured out from the training datasets by the algorithms (Oliverio & Poli-Neto, 2017). We can utilize clustering to find the classes inside a dataset. Other unsupervised learning algorithms include K-implies and Neural Networks.

Complete Chapter List

Search this Book:
Reset