There are three types of Machine Learning: Supervised Learning, Unsupervised Learning and Reinforcement Learning. Now in this article, we are going to talk about different models.
Before we jump into this, let's talk about how we split our data. We split our data into two parts 1) Training data 2) Test data.
From the training data, we generate a training model and from the test data, we test our training model. And then we determine the accuracy of our model.
There are different models which come under different learnings. As shown below,
Supervised Learning:
1) Classification
2) Regression
Unsupervised Learning:
1) Clustering
2) Dimensionality Reduction
3) Association
Reinforcement Learning:
1) Skill Acquisition
So, We know which model comes under which Learning. Now we each model in detail.
Regression Model:- Regression is basically used to identify a relationship between a dependent variable and an independent variable. So, what are the dependent variable and what is an independent variable? Let's understand with an example: One House with 1BHK has price 10000$, Second House with 2BHK has price 20000$ and 3rd House with 3BHK has price 30000$ So, guess what will be the price of the 4BHK house? Here we have information about price and house. We are predicting something depends upon previous variable value so, It is called regression.
Regression Model Algorithms:-
1) Linear Regression
2) Decision Trees
3) Random Forest
4) SVM
Classification Model:- Classification Model is based on predicting the target that has discrete values. For Example, There are lots of fruits in the bucket. Predicting the class of fruit is called classification. Another example is, Is the patient is suffered from fever or not? This is also the Example of a classification model.
Classification Model Algorithms:-
1) Logistic Regression
2) KNN
3) Decision Tree
4) SVM
5) Naive Bayes
[Note] In Supervised Learning, If the labeled output is continuous then we can use the Regression Model and If It is Categorical then we can use the Classification model.
Clustering Model:- This Model is used to find a structure in a collection of data. Clustering is organizing objects into groups whose Members are similar in some way. For Example, we have data abc, cab, bca So, it will give a result like aaa, bbb, ccc. Clustering algorithms can be categorized based on their cluster of the model. For Example, Connectivity-based clustering, Centroid-based Clustering, Distribution-based Clustering, Density-based Clustering.
Clustering Model Algorithms:-
1) PCA
2) K-Means
Association Model:- Association Model discover relationships in Large data sets. It is a rule-based machine learning model. It is used generally market-based analysis. For example, if you buy a Brush then apparently you also need paste for it. It works on the same rule to discover relations for large data sets.
Association Model Algorithms:-
1) Apriori
2) FP-Growth
There are two more terms you need to know is Overfitting and underfitting.
Overfitting:- The gap between the training error and testing error is large because the model is very complex.
Underfitting:- The training error is large because the model is too simple and can't capture the true complexity of the model.
We have covered pretty much of machine learning concepts till yet as a beginner needs to know. We will start the Next series on Neural networks and practical also. So, stay tuned with us for the next update. If you have any doubt then ask in the comment section or contact us on any social media. If you are here for the first time then check out our previous articles on Machine learning here.
0 Comments