Main Article Content
Abstract
With a rapid increase in advance and smart infrastructure in our daily lives, a huge amount of data generation occurs from various smart devices. This data usually falls under the category of big data as it is generated in large volumes with high speed at regular intervals. So, it becomes very challenging to represent and process such a large amount of data. Big data handling mainly involves data pre-processing and data processing. Finally, decision making is done on this processed data. Furthermore, data pre-processing plays a very crucial role in the efficient big data processing and data analytics. The data which is generated in large amount from various devices is of heterogeneous nature and uncertain. Big data pre-processing involves various steps, such as feature selection, feature extraction, data compression, reduction of duplicate and null values, and dimensionality reduction. High dimensionality of data is one of the main problem at this stage of data preprocessing to deal with. In this paper, a survey is done on various linear dimensionality reduction techniques.