Are you wondering what’s all this buzz about data science? The following videos will give you a good introduction to data science.
Use the links below to go to each video:
- Problems solved by data science
- Data Quality
- Question/Problem Quality
- Example of linear regression model
The first video is about the five questions/problems that data science can answer.
| Question | Problem | Example |
|
Is this A or B or C?
|
Classification | Will this customer leave or not? |
|
Is this weird?
|
Anomaly detection | Is this transaction typical when compared to other transactions from this credit card? |
|
How much/many?
|
Regression | What will be the sales for the following six months? |
| How is this organized? | Clustering | What are the groups of similar customers in this marketing database |
| What should I do next? | Reinforcement learning | In a self-driven car should I stop or accelerate if I see a yellow light. |
To have good data science models, data should be:
- Relevant to the problem you’re solving
- Connected (with the least number of missing values as possible)
- Accurate (with low bias)
- Enough to work with
Try to make the questions as precise as possible and to have adequate data for the problem you’re solving. For example, if your question is: what are going to be the sales for the next six months?, then you should have historical sales data.
In this video you’ll see how to make predictions of diamonds price based on the carats of them using a simple linear regression model.
Getting Started With Azure Machine Learning
One possible way of getting started in data science is to explore existing solutions to data science problems. The following video explains how to use existing solutions in Azure Machine Learning.
If you’re interested in learning more about data science, machine learning or R programming, please subscribe to the blog.