7 different types of Artificial Intelligent Algorithms

Share this post through

In this post, we are ae going to see Artificial Intelligence algorithms classification based on learning type and data type.

Classification AI based on Learning type

Artificial Intelligence(AI) algorithms are classified into 6 types based on learning type, results, and dataset type.


Image courtesy of Flickr

They are
  1. Machine Learning
  2. Deep Learning
  3. Reinforcement Learning
  4. Computer Vision
  5. Natural language Processing
  6. Natural Language Understanding

1. Machine Learning(ML)

Machine Learning is defined as a training machine or algorithm that can learn from experience without explicitly programming.

Machine Learning classified into three different types based on data characteristics. Before going to classification let us discuss types of data present in the world.
  1. Structured Data
  2. Un structured Data
  3. Semi-Structured Data

1. Structured Data

  Structured data is associated with data that have labels for every value or character in the dataset.
Examples of these types of data are CSV files, images with labels..etc.It is very easy and interesting to deal with this type of data.

2. Unstructured Data

 Unstructured data is not having any labels or codes for data. 80% of world data is unstructured. It very hard to handle this type of data.
Examples of Unstructured data are photos, videos, documents, and audio files.

3. Semi-Structured data

This is a combination of Structured Data and Unstructured Data. Examples of this type of data are JSON and XML files.

Classification of Machine Learning

     Machine Learning classified into 4 types
  1. Supervised earning
  2. Un supervised Learning
  3. Semi-supervised learning
  4. Association rule learning                

2. Deep Learning

This is a back-propagation based learning algorithm, that corrects errors automatically. Deep Learning again classified into 7 different types of algorithms.
  1. Artificial Neural Networks(ANN)
  2. Convolutional Neural Networks(CNN)
  3. Recurrent Neural networks(RNN)
  4. Generative Adverse Neural Networks(GAN)
  5. Self Organizing Maps(SOM)
  6. Boltzmann Machines
  7. Auto-Encoders

3. Reinforcement Learning

 Reinforcement Learning is rule-based learning, for every task algorithms follow a pre-defined process. Let me explain in detail using Ludo Game.

In Ludo Game, there is an option like play with Computer. That is an AI-based algorithm that follows certain rules and moves automatically according to that pre-defined actions.
Reinforcement Learning classified into four different algorithms. They are
  1. Q Learning
  2. Deep Q Learning
  3. Deep Convolutional Neural Networks
  4. A3C Algorithm

4. Computer Vision(CV)

This is the eye for Artificial Intelligence. This is used for image and video analysis. Majorly Computer vision has 4 sub-categories. They are
  1. Face Detection
  2. Motion Detection
  3. Object Detection
  4. Wireless Action Implementation

5. Natural Language Processing(NLP)

    Natural Language Processing dead with Language processing, understanding, sentiment analysis. The sub-categories of Natural Language Processing are
  1. Chat Bot Development
  2. Sentiment Analysis
  3. Emotion Extraction
  4. Summarization
  5. TF-IDF Model
  6. Word2vec Analytics
  7. Bag of Words  Model
  8. LSA Model

6. Natural Language Understanding(NLU)

  This is a subcategory of NLP. The sub-categories of NLU are

  1. Speech Recognition
  2. Voice Assistants
  3. Text to Speech
  4. Question Answering
  5. Machine Translation

7. Transfer Learning

 Transfer learning does unknown tasks by using Knowledge and representation of another task. The Main sub-categories of Transfer learning are
  1. LeNet
  2. VGG16
  3. GoogLeNet
  4. ResNet
  5. 3-4 CNN
So far in this post, I have almost all AI algorithm categories based on learning flow and data type.

Thank you