Artificial Intelligence is an umbrella term and describes the broad approach of using machines to imitate intelligent human behavior in order to solve problems.
Machine Learning is a technology used to achieve Artificial Intelligence. For example, if one were to develop an algorithm to detect fraud in financial data, this would be typical AI. If this algorithm still learns itself and also recognizes new facts, it would be called ML. Deep Learning is the further development of Machine Learning. The technology makes use of so-called neural networks (similar to how the human brain works) or artificial neural networks.
“People worry that computers will get too smart and take over the world, but the real problem is that they're too stupid and they've already taken over the world." ― Pedro Domingos
Tools and conclusive analysis
Tools What are the tools like programming languages, services and software you can use for data science? Programming Languages — Famous programming languages are:
- R
- Python
But also C# and SQL are very common for data science tasks. Services — Here you will find a lot especially in the clouds of the big three like Google, AWS and Azure. Often used services are e.g. image and video recognition, translation, NLP and many more. Software — Here will find also many useful software. Free software like:
- Anaconda
- Jupyter Notebook
R Studio
and also software you have to pay for like:
alteryx
- MS Power BI
- Tableau
- Methods
Besides certain tools you will also need a data science process. Here, the most famous process to develop AI is the CRISP model.
CRISP-DM stands for “Cross Industry Standard Process for Data Mining”. It is a standardized process model that can be used for data mining in order to search data stocks for patterns, trends and correlations. For this purpose, the standard defines six different phases that are to be run through once or several times. The model can be used across many disciplines like data science, business intelligence and engineering . Supervised, Unsupervised and Reinforcement Learning Supervised Learning: Here, the algorithms are defined on the basis of specific examples. An attempt is made to find the solution for other similar problems by generalizing a solution. Supervised learning can be used, for example, to predict customer churn. Popular examples are classifier and regression analysis . Unsupervised Learning: Here, the algorithms are processed with arbitrary examples. The goal here is to identify a structure within the data set. The most important method is clustering.
Reinforcement Learning: Reinforcement learning or reinforcement learning stands for a set of machine learning methods in which an agent autonomously learns a strategy to maximize rewards received. One of the most famous algorithm as an examples is the Monte Carlo algorithm. Conclusion
This is in short what you have to know about Artificial Intelligence. To dive deeper you can use the sources below or just google a bit. Famous platforms to start learning and gaining certificates are for example Udemy, Data Camp or Udacity.