kindly Read through abstracts,copy edith and give to your surpervisor....
PROJECT TOPIC: BIG DATA CLUSTERING USING HADOOP ALGORITHM
Department: Computer Science
FORMART: MS WORD
PAGES: 72 pages, abstract, chapter 1-5 , APENDIX A source code and APENDIX B output, well reserached and supervise
Get Complete Work
Abstract In the last few years there has been voluminous increase in the storing and processing of data, which require convincing speed and also requirement of storage space. Big data is seen as large, diverse and complex data sets, when dealing with massive data sorting.In this work, big data clustering system using hadoopwill be developed. Hadoopis the most commonly used parallel clustering algorithm which has efficient execution time frame on big data clustering.This work centres on clustering staff record within an organization, the proposed system adopted object oriented Analysis and design (OOAD) methodology and will be implemented using Visual Studio IDE, the proposed system will be focused on Big Data document clustering by providing less time to cluster single data based on similarity and parallel from large data set especially organizations that deals with high data volumes.This work will be beneficial to organizations with large work force and other researchers who wish to improve on hadoop algorithm on big data.
CHAPTER ONE INTRODUCTION 1.1 Background to the Study The development of Internet technologies and social media revolution has increased the speed, and the amount of data produced on daily bases. Companies and business organisations have started to collect more data than they know what to do with. Hence,Big Data has rapidly increased in different application areas and tends to dominate future technologies, Zen, Y.(2000). Big data can be defined as high-volume, high velocity and high-variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. This type of huge volume of data of course brings several challenges. Several challenges include analysis, retrieval, search, storage, sharing, visualization, transfer and security of data. Data generation capacity has never been so big, every day 2.5 trillion bytes of data are generated and 90% of produced data in the world so far were produced in the last two years. However, the data are not in a meaningful form they are in the raw form. It is necessary to process them to explore patterns, hence transforming them to meaningful information. If they are used in an accurate manner,Big Data have the ability to give results on these massive data piles, which shed light to future technologies Bay, E.(2012). Doing so, data mining methodologies will play an important role to explore data. Data mining is defined as the science of extracting useful information from large datasets or databases. This area combines several disciplines such as statistics, machine learning, artificial intelligence, pattern recognition, and data management.The idea is to build computer programs that examine the databases automatically, search for rules or patterns in order to make accurate predictions about future Data. Since actual data are usually imperfect, it is expected that there will be problems, exceptions to each rule. Therefore, the algorithms must be robust enough to adapt to the imperfect data and extract regularities that are inaccurate, but useful. One of the most important problem sources in Big Data and data mining is still the heterogeneity of data structure and data resources when gathering data. The term Big Data refers to large-scale information management and analysis technology that exceed the capability of traditional data processing technologies. The incorporation of Big Data is changing Business Intelligence and Analytics by providing new tools and opportunities for leveraging large quantities of structured and unstructured data clustering systems like k-means, hadoop according to Modha, D.(2014). Big Data is notable not because of its size, but because of it’s relationally to other data. Due to the methods used to store the data, Big Data is fundamentally networked (threaded with connections). But these connections are not useful directly. The actual value comes from the patterns that can be derived from the related pieces of data about an individual, about individuals in relation to others, about groups of people, or simply about the structure of information itself. Besides this, Big Data has enormous volume, high velocity, much variety and variation. These features of Big Data present the main challenges in analyzing big data which are Efficient and effective handling of large data, Processing time and accuracy of results trade off and Filtering important and relevant data from all the data collected Malewicz, F.(2009). Traditionally, large data is handled through numerous data mining techniques. Recognizing patterns among data must borrow ideas from Machine learning algorithms. Thus, big data analysis needs fusion of techniques for data mining with those of machine learning. Hadoop is a framework that allows for the distributed processing of large data sets across clusters of using simple programming model like map reduces. It is designed to speed up from single process to hundreds of or thousands of data storage and computation.