How To Become a Data Scientist? Part 1
First, What exactly data science is and how it has become so popular these days? I am going to give some ideas about the tools, skills, education you need to become a data scientist.
Data Science is a totally broad subject involving various sets of skills from leadership to domain knowledge and programming language.
In Part 1, I will be explaining the meaning of data science and process.
In Simpler Terms, Data Scientists are the chef ( Food Scientist)
The word chef is a French word which means ‘head of the kitchen’. Any person can be a chef by graduating from cooking school or learn from an experienced chef while working with them. A chef is a trained professional cook who is proficient in all aspects of food preparation, often focusing on a particular cuisine.
Data scientist are like a chef who gave you delicious yummy food. The chef is the artist who designs the recipe for the food and data scientist are the one who designs model for our specific use case.
I would like to call a chef as a highly-skilled professional in the food industry who is responsible to come up with recipes for delicious food. Once the successful recipe is prepared, which then can be implemented into a chain of restaurants in large scale to generate revenue.
Now, you can relate data scientist as a food scientist, who are responsible to design model to solve a business problem.
Fig1. shows how similar the Data Scientist and chefs are related. Let’s discuss some of the process involved in building a model or recipe for the specific application.
Problem: Problem statement should be clearly defined before moving to the data preparation model. For example, if somebody gives you all the ingredients, vegetable, to cook the dinner for you at home. First, you should identify the personal taste, who you going to serve, and you should know if the person wants spicy, sweet, or half-cooked. If you know all the details of the person, you will prepare the delicious food. Identifying personal taste is similar to defining the problem statement.
Data Preparation: Data scientists are given with different data and should come up with a model that should benefit the company for the long run.
Data scientist are required to work with structured and unstructured data in the real world. As seen below, food scientist will start buying ingredients for the food and then do all the cutting, cleaning the ingredients before preparing the food. A similar method applies to data scientist, where they will collect data from different sources and then clean the data for model building step.
Model Building: Model building step is similar to the chef, who is ready to prepare the perfect dish for the customer or individual. For this, a chef should know about the quantity of salty, spicy and all other ingredients to prepare the delicious food. Once, all the quantity are identified, the chef can launch this recipe to all the restaurants. While, in the data science field, this is exactly finding the right algorithm and doing all the hyperparameter tuning.
Evaluation: Are you 100 per cent sure that your recipe is perfect to prepare a good dish. For this, you will ask your family members to taste the dish. To be more confident, you will invite your friends or neighbours to taste your dish. If everybody love, then you will launch this recipe to the restaurant. In this step, the model is evaluated to identify how the model generalizes with the unseen data usually in test or validation data sets. There are a lot of different metrics to evaluate the model too.
Deployment: Once the model is finalized with acceptance criteria, it will be deployed for the production. Once the yummy recipe is developed, it will be used for the commercial purpose as seen in below figure.
So, data scientist are like a chef who invests time and knowledge on collecting, cleaning data to solve the real-world problem.
This is the slight introduction to the data science and in the next part, I will be talking about the essentials skills needed in data scientist field. Till then Keep Learning and exploring.