Before you decide what to wear in the morning, you collect various types of data. Data analysis is the process of inspection, cleaning, modification, and modeling of data with useful information to find useful information, informing findings and making decisions. There are many facets and approaches in data analysis, in which various techniques, science, and social science domains are used under different names.

Data Mining

Data mining is a specialized data analysis technique that focuses on modeling and searching for knowledge rather than fully descriptive purposes, while business intelligence covers data analysis, which mainly focuses on business information, heavy on aggregation depends on. In statistical applications, data analysis can be divided into descriptive data, test data analysis (EDA), and vertebrate data analysis (CDA). EDA is focused on exploring new features in data, while the CDA focuses on confirmation or misconception of existing estimates. The estimated analysis focuses on applications of statistical models for prediction forecasting or classification, whereas text analytics implements statistical, linguistic and structural techniques to remove and classify information from text sources, a species of structured data may apply. All of the above data are varieties of analysis.

Analysis and Interpretation

 By analyzing general trends in analyzing data and by pointing out the differences and similarities between data points, the reader is able to understand the data. Interpretation is related to data that considers data measurable, searches for relationships between various educational objectives, qualifications, expansions, images, and assessments. Questions about questions after analysis and interpretation: What is the data about your students’ subject matter, research skills, writing and speaking?


Qualified research involves the meaning and meaning of the data collected, and through which emerging information is applicable to customer’s problems. This data often takes the form of archives of group discussions and interviews, but it is not limited to this. Through the process of amendment and immersion in data, and through complex activities of the structure, remodeling or otherwise exploring, the researcher seeks patterns and insights about major research issues and uses the brief address of the customer to address them.

There are many steps described below can be categorized. Stages are repeated, from the later steps that may have additional functions in the first step in that reaction.

Data requirements

Data is required as an input for analysis, which is specified based on the analysis or requirements that guide the customer (who will use the finished product of the analysis). The general type of entity on which the data will be collected is known as an experimental unit (for example, person or population). Regarding the population, specific variables (for example, age and income) can be specified and received. The data can be numeric or clear (i.e., a text label for numbers).

Data collection

Data is collected from different sources. Analysts can be informed by analysts such as information technology personnel within an organization. The data can also be collected from the sensor in the environment, such as traffic cameras, satellites, recording devices, etc. Interviews, downloads from online sources, or reading documents can also be obtained.

Data processing

Intelligence information is used to convert raw materials into working intelligence or knowledge, which are conceptually similar to phases in data analysis. The data received at the beginning must be processed or organized for analysis. For example, they can be included in the spreadsheet or statistical software to keep data in rows and columns in table format (i.e., structured data) for further analysis.

Data cleaning

Once processed and organized, the data may be incomplete, duplicate or error. The need to clean data is generated by the problems in such a way that the data is recorded and stored. Data cleaning is the process of preventing and improving these errors. Common tasks include recording matches, identifying data incorrectness, the overall quality of existing data, deduplication, and column division. Such data problems can also be identified through various analytical techniques. For example, with the financial information, comparisons of the total for special variables can be considered comparatively reliable compared to individual published numbers. Abnormal quantities above or below the predefined threshold can also be reviewed. There are many kinds of data cleaning which depends on the data number like email number, email address, employer etc. Quantitative data methods for external identification can be used to get rid of the data entered incorrectly. Text data can be used to spell checkers to reduce the number of incorrect data words, but it is difficult to say whether the words themselves are true or not.