EP 2. APPROACHES TO MACHINE LEARNING:

 EP2. Approaches to Machine Learning:

Today Lets get started with the approaches in ML.
  • Machine Learning techniques are required to improve the accuracy of predictive models. There are actually different techniques used on different data and analysis stuff.
  • SUPERVISED LEARNING: Supervised Learning typically starts with a well-addressed data and a certain understanding of how the data is classified. We actually give the well-structured data and apply machine learning algorithm we will get the output. So indirectly, we are telling the machine how the output will be like and what should we get actually from that. We actually tell the machine how to get the predictions for us. 
  • The above picture represents the supervised learning.
  • There are 2 types of Supervised Learning: 1>CLASSIFICATION     2>REGRESSION

  • For example when you consider a patient is to verify if he is actually suffering from cancer or not: we actually use the classification technique so the machine will predict that whether the person is benign(Not harmful) and Malignant(harmful). This type of supervised learning algorithm is used many times in our day and its very important one! The output is discrete-valued output which is like 0 or 1 etc.



Here it is divided into 2 parts where the RED dots represent benign and the BLUE dots represent malignant.
  • For example when you want to find the housing prices of an area then we will get lots of results i.e it is based on many factors and we actually get is continuous-valued output like when you actually consider an area and it varies based on the luxuries provided with that and you also get a different money call from a broker and it will reduce if you buy it from your friend. So we use quadratic functions and many others to get the desired price and the customer, as well as seller, should be satisfied with the decision they made so that's why we say regression is continuous-valued output!
  • The below graph represents the budget and sales and based on the data we have drawn a hypothesis(red colour line) to get our desired output.



AUTHOR: Kaushik Tummalapalli
So In the next eposide, we will be looking closely about unsupervised learning.  Have a Nice Day!
Thank You,



Comments

Post a Comment

Popular posts from this blog

"ABOUT MICROSOFT STUDENT PATNER'S PROGRAM(MSP'S)"

WHAT DOES AN MSP DO? Hello Guys, Im kaushik Tummalapalli and recently I got selected as an MSP for 2020-2021 and MSP actually stands for Microsoft Student Partner. Some Basic Glance about MSP: Microsoft Student Partners (MSPs)  are student technology leaders, empowered to build Microsoft communities on their campus and share their deep knowledge and passion for technology with their fellow classmates. Spanning more than 100 countries around the world, MSPs receive: Leadership Experience  -  Host fun workshops, run hackathons, and give demos on campus to grow a community of students. Resume Building  - Enhance their promotional skills, add to their professional technology experience, and work alongside Microsoft professionals. Exclusive Access  - Receive insider training, exposure to career opportunities, and access to the latest technology events. Some Important tips to get selected in this program as an MSP: 1> If you have passion in technology you have a greater
Read more

EP: TERMINAL COMMANDS

Terminal Commands :  Here in this module i'll be going through some basic commands that may be very useful when   we are using a command prompt/Terminal or some other computer system like linux etc. So I'm currently using a mac book air version where I have got a pre-installed terminal otherwise you can find it in your files in drives/files in your computer system(ALERT: I'm using MAC BOOK AIR TERMINAL) 1. So the first command is cd ( which actually stands for change directory 2. ls (it will get the total list in that directory) 3. ls -l (we get the list of  everything in a list) 4. clear (This command is used to clear all the commands in the terminal) 5. pwd (This command is very much used to know in which path we are in or the location we are moving) 6. cd ../ (This command is used to go back one step or in other words go back to the previous path or directory) 7. mkdir (We make a directory which means to create a folder on our current directory or pat
Read more

EP 4. MACHINE LEARNING CYCLE

EP 4. MACHINE LEARNING CYCLE: Creating a machine learning application or implementing a machine learning algorithm to some data is an iterative process . We can't simply develop a machine learning algorithm and just leave it because data changes , preferences emerge. When we train our machine learning model with a machine learning algorithm, for each and every algorithm there is different accuracy. So we should choose which one fits correctly in terms of all metrics like f-square, mean absolute error, p-value etc. The machine learning cycle is continous cycle  and choosing which algorithm fits is only one of the step in the cycle.The steps in the machine learning cycle are as follows: 1> Identify the data : First of all when we have a dataset full of data and features(columns) we have to select which feature we want because it will be an underfit when we try to predict. We can do feature engineering when we want to add any features.Then the next step is to observe how the
Read more