METHODOLOGY FOR DATA SCIENCE

DATA SCIENCE METHDOLOGY

Bussiness Approach :

Understanding:

The problem that we are trying to solve.

Analytical Approach?:

How to use data to answer the question.

Case study: ( Implementation of concepts)

• What is the best way to solve ?
• What are the goals and objectives

Analytical Approach :

Pick most appropriate question.

Second stage?of data science methodology

Descriptive

• Current status

Diagnostic(Statistical Analysis)

• What had happened?
• Why is happening?

Predictive (Forecasting)

• What if these trends continue?
• What will happen next?

Prescriptive

• What’s the solution?

Questions under considerstion:

If the question is to determine the probability of an action .

• Use predictive model

If the question is to show relationship

• Use descriptive model

If the question is to require answer?yes/no

• Use a classification model

Will Machine Learning be utilized??

• Learning without being explicity programmed
• Identify relationship
• Uses clustering assosiaction approaches.

Case study : (Implementation of concepts)

• Decision tree classification .

Predictive model

To predict an outcome.

Decision tree classification

• Categorical outcome
• Likelihood of classification outcome
• Easy to understand and apply

Why is the bussiness understanding stage important?

It helps to clearify the goal of the entity regarding the question to understand the problem and tries to solve the problem.

Why is the analytic approach stage important?

Because it helps to identify what type of patterns will be needed to address the question most effectively.

Conclusion:

In this article, you have learned:

• The need to understand and prioritize the business goal.
• The way stakeholder support a project.
• The importance of selecting the right model.
• When we will use a predictive, descriptive, or classification model.

DATA UNDERSTANDING:

Data understanding encompasses all activites related to datasets.

• What does it means??
• To prepare or clean?

Case Study?:

Understanding the Data

• Descriptive statistics
• Univarite statistics
• Pairwise correlation
• Histograms

Case quality :

Viewing data

Data Quality:

• Missing values
• Invalid or misleading values.

DATA PREPARATION:

Data preparation is a very important step . We have to remove unwanted elements and keep those elements that is useful for us.

• The importance of descriptive statistics.
• How to manage missing invalid or misleading data.
• The need to clean data and sometimes transform it.
• The consequences of bad data for the model.
• Data understanding is iterative; you learn more about your data the more you study it.

MODELING TO EVALUATION:

Model:

In what way can be data visualized to get the answer that is required?

Evaluation:

Does the model used really to answer the question ?

DEPLOYMENT:

The solution will be given to the stackholders.