Data analysis refers to the process of cleaning, inspecting, modelling, and transforming data with the sole aim of gaining useful details, which support decision making. This term has got different approaches and features which include a variety of techniques under different names in many fields like science, business, social science, etc.
Data mining is a form of data analysis which deals with knowledge discovery and modelling for the sake of predictive purposes rather than descriptive ones. Business intelligence makes use of data analysis for the aggregation of business information. In statistics, data analysis is divided into exploratory data analysis or EDA, descriptive statistics, and confirmatory data analysis or CDA. CDA is used for falsifying or confirming the already existing hypothesis, while EDA is used for finding out new features in the concerned data. There are two other varieties of data analysis namely predictive analytics and text analytics. The former makes use of structural or statistical models for classification or predictive forecasting while the latter utilizes linguistic, structural, and statistical techniques to gain as well as segregate information from unstructured data.
Data analysis is a process, which includes several phases. Data cleaning is the first stage of a data analysis, and refers to the prevention and correction of errors in the methods through which data is entered as well as stored. There are phases in data analysis like initial analysis and main analysis. The initial phase includes four questions regarding the quality of data, quality of measurements, initial transformations, and finally relevance of the study in satisfying the intentions of a particular research design. In the main data analysis, either confirmatory or exploratory approach is adopted. In the confirmatory approach, the hypothesis related to the data is tested, while in the exploratory approach, no such hypothesis is addressed.
In short, data analysis involves a series of procedures.