Every datum of a Data Analyst!

Yup! I am glad you guys are interested in this article on this title. As a beginner, self-learning person I wrote this article on the knowledge I gained.

The world of data is quite the same as a black hole. Every datum has its own Power/Value. Today in the fast and furious world the concept. The budding term Data Science, Big Data, Data analytics is rapidly updating its structure. Every piece of data is used in any means of time. To get various predictions, forecasts or to make a decision, to realize of update the deep analytics of the previous or the market one is essential. By this means data analytics follows the SIX phases as, ask, prepare, process, analyze, share, act.

ASK: This is the first process of data analytics. In this phase the recognization of current problem/situation, organizing available information, opportunities and identification option. This phase can be also said to be questioning what?. SMART questions pave a vital role. ( Specific Measurable Action-oriented Relevant Time-bound.). Fairness is important while asking questions. Data Analyst communicates with the Stakeholder. Stakeholders are those who need to analyze the data. They invest time and resources to analyze the data to make data-driven decisions.

" Communication is a Key "

PREPARE: Data generation, collection, storage, and data management are the key factors talked to be in this process. As the result of the first phase, we get sufficient tasks to be done. Our data may contain discrete, continuous, nominal, internal, external, structured, and unstructured data. We handle huge amounts of data, so sometimes we are supposed to do sampling. In such cases, the data must not be biaesd. A biased analysis leads to misleading the stakeholder.

The data we handle must be fairful for all data ethics as ownership, Transaction, consent, currency, privacy, and openness. Metadata data about the data should be maintained. In the case of internal data, we must care for the protection of the data since the data is private.

PROCESS AND ANALYSE: The data we obtained so far is good but might contain incorrect, unwanted information. So the processing is done. Sorting, Filtering, transforming data in a useful format for analyzing and giving of reliable infrastructure is done in this process.

Quality of data determines quality of analysis

In beginning, the R-Programming is used for analyzing the data. Here you may ask me why not now...! Yeah I have a valid point with Python. I personally recommend Python for handling the data. The modules and library in python make the analyzing phase very simple than the ancient ones.

DATA VISUALIZATION: This is the phase that is the most satisfied and my favorite part of data. This gives a comparative structure to the data we are working with. The visuals of the data are represented as a pie chart, line chart, graphs, histograms, etc,... This visual representation gives the correlations among data. Some tools like Python(Matplotlib), Tableau, etc... can help in this part. This phase makes a data analyst convey the part of data differences to the stakeholder.

ACT: The action that the organization/stakeholder is discussed with all factors. Clearly share, discussed data-driven-decision is taken as the final action. In the case of automating (AI) the action by the data's decision is to be implemented, we move on to Statistics and Machine Learning that's the story for another article!