By : Suraj Joshi | On : May 15, 2017
Here is a quick write-up on some of the common terms used in predictive analytics
Artificial intelligence is a study of intelligent agents. AI was established over the thought that human intelligence "can be so precisely described that a machine can be made to simulate it".
An artificially intelligent agent is an agent with some goal and it understands its surroundings and takes suitable actions that increase its odds of reaching its goal. The term AI is often used when computer mimics the cognitive capabilities of a human brain such as learning and problem-solving.
Skills which are being considered as an AI involve Human speech recognition, self-driving cars, CDNs, interpreting complex data etc.
Machine learning is one of the sub-domains of computer science. It provides a capacity to computers which enables them to learn without being programmed. Machine learning has been developed out of the subjects like pattern recognition and computational learning theory in artificial intelligence.
It is basically a study of building algorithms which analyze the data to learn the pattern and then make predictions out of it. Which makes it purely data driven and disregard the need to write static program instructions.
Machine learning is applied to a wide range of computing task where the application of static programs is not worthwhile. Some examples include spam filtering, OCR, Search engine ranking, Computer vision, Image recognition etc.
Artificial neural networks is a computational model which is used in the field of machine learning. An artificial neural network is a collection of lots of connected sample units called artificial neuron which forms a network and mimics the behavior of neural network in the human brain.
These networks are trained by providing them sample data and not by writing static programming codes. It has been employed to solve the wide range of tasks like computer vision, speech recognition and other complex things which are hard to solve by using explicit programming methods.
Deep learning is a field of machine learning which uses the techniques of a human brain to learn and understand the data. It enables the machine to learn layer by layer with different levels of abstractions. Each layer learns and processes the data and passes on the information to the next layer which processes it further.
Until now we didn't have enough data processing power to train the machine to learn but with the use of GPUs computational power of machines have become robust. The process of deep learning has been accelerated immensely with the help of parallel computing power facilitated by the GPUs.
It is being used in a variety of fields like Natural language processing, facial recognition etc.