Introduction of data science

What is data science?

Data science is the field where we extract meaningful data from raw data, which we can use in various works like creating machine learning model, AI, Business analysis, text and advanced Image recognition, banking, finance, manufacturing, transport, e-commerce, education, Healthcare, recommendation system, gaming, security, etc. So we can say that, the area of data science implementation is huge.

Why data science is important?

Per day approximate more than trillions of data are generated. To grow a company or get more profit or solve the daily problems, we need to use those data. To use those data, data science works.

As I told you, in data science, we extract meaningful data from raw data.
So here trillions of data are the raw data and a data scientist extract meaningful data from those trillions of data, so that we can work with those data.

Nowadays every company understood that, they have to use the data to grow. So that they are using data science to complete the work. So the opportunity of a data scientist is huge.

How the work is done of data science?

Data science life cycle image

Step 1:
First, we have to identify the problem and then have to understand the problem.
Step 2:
After understanding the problem, according to the problem we have to get the data.
Step 3:
After getting the data, we have to process and clean the data. Look, at first we have raw data. In this data, we can have missing value, noisy data, some data which are not that much important, sometimes we need to create new data according to the existing data to get better result. So we need to process and clean the data.
Step 4:
After the previous process, now we have the clean and preprocess data. Now we have to analyze these clean data. To analyze, we can create graph and chart, some statistical analysis, find the relation between data, etc.
Step 5:
Now we have to create a model. We can create machine learning model, deep learning model, etc. In that model we will feed those cleaned data that we have. After feeding the data, our model will train itself using those data. After training we have to evaluate the trained model.
Step 6:
Now we have to check the model working. To check we can feed some new data and can check that the model is giving right prediction or not.
Step 7:
Now we have to deploy the model to solve the problem that we got.

So this way the works happen in data science.

What are skills you should have to become a data scientist?

Step 1: Python or R
Learn a programming language between Python or R. Most popular is Python. Most of the data scientist are using Python in their daily data science work. It is popular because the syntax is easy and has too many popular libraries or packages to solve data science problem. So I will recommend you to learn python but you can learn any.
Learn more about Python
Step 2: mathematics
Learn math. You need to learn mathematics.
The mathematics topics are linear algebra, matrix, vector, probability and statistics.

You have to also learn the Matlab. Matlab is a programming language and numeric computing platform.
Using Matlab :
1. You can easily implement and test the algorithms.
2. You can easily develop the computational codes.
3. You can use a large database of built in algorithms.
4. You can process still images.
5. You can easily create simulation videos.
6. You can easily do symbolic computation.
7. You can easily perform extensive data analysis and visualization.
7. You can easily develop application with graphics user interface.
Step 3: Numpy
Learn numpy. It is also very popular library. Using this library you can perform matrix related work, generate matrix, can work with 1,2 or 3 and more dimentional array, etc.
Learn more about Numpy
Step 4: Pandas
Learn pandas. It is very popular and powerful library of python. You can used this library to get data, clean data, process the data and various works.
Learn more about Pandas
Step 5: Matplotlib
It is library of python. Using Matplotlib you can visualize the data by creating graphs & charts.
Learn more about Matplotlib
Step 6: Seaborn
It is library of python. Using Seaborn you can visualize the data by creating graphs & charts. The difference between Matplotlib and Seaborn is, Seaborn offers you more graph & charts with more controls and beautiful graphics compare to Matplotlib.
Learn more about Seaborn
Step 7: Machine Learning
Now you have to learn machine learning. Using machine lerning you will create your data science model.
Learn more about Machine Learning
Step 8: Github
Github is a place where you can upload your code to share with others and also get code of various persons. It is very important thing. If you are working in a field of computer science then you must learn github.
Learn more about Github
Step 9: OpenCV
It is a very powerful and popular python library. It is used to work with image and video. If your working with images and videos then you will need these library.
Learn more about OpenCV
Step 10: Deep Learning
Deep learning is also used to train your model. For small data machine learning works well but for big data deep learning works well.
Learn more about Deep Learning
Step 10: Web scraping
Using web scraping techniques you can get or collect data from web pages.
Learn more about web scraping
Step 11: SQL
In data science, you will work with the data base. You will use the data base to maintain data base data, get data from data base or upload data into the data base and various works. So a data scientist should also learn about the data base.
Learn more about SQL
Step 12: MongoDB
In data science, you will work with the data base. You will use the data base to maintain data base data, get data from data base or upload data in the data base and various works. So a data scientist should also learn about the data base.
Learn more about MongoDB
Step 13: Excel
In data science you will work with data. Excel is a very powerful software to work with data. So you have to learn about excel also.
Step 14: Docker
Docker is used deploy project from one machine to another machine.
Learn more about Docker
Step 15: Power BI or Tableau
These are visualization softwares. These softwares will help you to create dashboard of graphs and charts to analyze the data and represent the analyze data. You can learn any between Power BI and Tableau.
Learn more about Power BI
Learn more about Tableau