Question

Chapter 9 mention 5 categories of artificial intelligence. Use the internet and research 3 types of artificial intelligence not mention in Chapter 9. Explain how each works and provide an example on how each are being used.

PLEASE DON'T COPY SOMEONE ELSE'S ANSWER ON CHEGG BECAUSE I ALREADY SAW THEM.

Answer 1

Well Although the context of the problem is not very clear as i don't know what actully chapter 9 contains and what categories doest that actually talk about, but an per my knowledge the Artificial Intellengence or Machine Learning mainly have 3 categories divided by how they approch the problem statement although according to some people they also say there are 4 categories and they are

1. Supervised Learning
2. Unsupervised Learning
3. Renforcement Learning
4. Semi-Supervised Learning (Can vary how one defines it)

let's see what they actually are and some applications of it

1. Supervised learning is an approch where you have a defined output. which is already been given it is very similar to teaching machines or algorithms by saying them if some fruit is round and medium in size the probability is that fruit could be orange means you actully try to find some features to find the patterns as human we also finds the pattern in things like same as orange we have judgement it is orange because of its color shape size or the pulp or fuffyness you feel where on the same hand if same circle is larger in size and green color with hard surface it could be water mellon, so similarly the machine or algorithm tries to find the patters in data to define it in it's own language and it is called supervised learning as you data is already labeled. Some useful Applications
   • Trend Analysis (Stock Market Predictions)
   • Predicting housing prices (Regression Problem)
   • Classifying weather a person has a disease or not
   • email - Spam Classification
2. Now let's assume the same above case but in that we are not telling the algorithm in prior that what actually that fruit is, So in that case some clustering techniques comes into the picture where algorithm tries to create groups elements which are similar in nature for example if it finding patters of shape and sizes it will create two different groups for oranges(Sphere) and different for banana(Cylinderical) and give them there own labels as we have not told him what actually that fruit is so it named them something like cluster or group 1 and 2. Some application
   • Anomly Detection or Fraud Detection (as the user behaviour is going to change distrastically so machine will able to understand the something is wrong)
   • Sentiment Analysis (Judging the mood or sentiment of the user by the choice of his words like hate, worst impose negative sentiment and love wonderfull positive sentiments)
   • Recommendation Systems (You might be wondering how netflix is sugessting you different shows or videos it is recommendation engine or systems are working at backend to sugessest the things you might like)
3. The major is renforcement learning it is kind of try and run method it is like a baby learning how to walk each time he/she walks a step properly you give them some rewards and everytime they fall you punishes them. So, actully in code you just do +1 for reward and 0 or -1 for fall or wrong move by the rewards system or algorithm understands that this actully is the right thing to do and keep following the thing unless its not getting more rewards from that, Practical Applications
   • AI based game developments (Open AI Dota 2 world champion model is based on it)
   • Self Driving car

Other than that there is something called Neural Networks which follows the same footsteps but are advanced algorithm that are trying to memic the human brain functionality