Skip to main content

Introduction to Machine Learning

Salmo Aliko .
today I'll write about Machine Learning , it is interesting topic. I put assumption I'm talking to computer since students.
first of all , many of us dreamed to make robot, auto detection for missiles and treat it automatically and many awesome dreams, but we didn't know how achieve it.

let me start with the definition and mentioning some of ML -Machine Learning - applications .
What is Machine Learning? 
 is the ability to learn without being explicitly programmed 
 and if we like mentioning more formal definition it is 
 A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.
application of ML :

  • Computer Vision.
  • Natural Language Processing.
  • Game playing.
  • Market segmentation.
  • Medical Diagnosis.  
  • Stock Market Analysis.
  • Search Engines.
  • Robotics.
One word describing this applications just "awesome".

Machine Learning can be classified into 3 main categories .

  1. Supervised Learning.
  2. Unsupervised Learning.
  3. Reinforcement Learning.
lets talk in brief about each category :
  1. Supervised learning : we have a data set labeled i.e. ( input and correct output ) using this data set given new data not labeled  expected the output is correct for example neural networks.
  2. Unsupervised Learning : you are given a data set not labeled you find appropriate structure then given another data you should output good estimation for example market segmentation.
  3. Reinforcement Learning : you are given data set using Learning algorithms you can judge either the output is right or is wrong and every time the algorithm will learn for example :
  • robot driving car at turning the robot made the angle very acute so he made an accident the learning algorithm will know this is wrong at the next time the angle will not be acute, after many iterations he will know that the right angle is obtuse "Eng Ali Al kahki ".
after that I hope I gave you good intuition about Machine Learning.

if you like this Article share it.

I owned by  a great Thanks to Eng Ali Al Kahki, the person who explained this concepts and the best prof.  I have ever seen prof. Andrew Ng.

Thanks for reading.

Ahmed Ghazey.

Comments

Post a Comment

Popular posts from this blog

DevExpress C# Chart Tutorial

Hello , this blog post about Charting in DevExpress , I faced many problems I visited hundreds of pages to solve this problems , actually it was the first time I use DevExpress Charts, Work is not like school or university , in school if you don't know thing you can learn it any time you want , in work you have to do today task yesterday, I like to share my knowledge and skills with you , I hope this Post help you . // clear series on chart  chart.Series.Clear(); // clear Titles of chart chart.Titles.Clear(); // legend alignment  chart.Legend.AlignmentHorizontal = LegendAlignmentHorizontal.Center; chart.Legend.AlignmentVertical = LegendAlignmentVertical.Bottom; chart.Legend.Direction = LegendDirection.LeftToRight; // creating series // series view Type example Line DevExpress.XtraCharts.Series series = new DevExpress.XtraCharts.Series("series name ", ViewType.Line); // add title  DevExpress.XtraCharts.ChartTitle chtTitle = new DevExpress.XtraCharts.Cha...
3 comments
Read more

Designing Entity Relationship Diagram (ERD)

Al Salmo Aliko . Hello, in this post I'll discuss with you how to design good Entity Relationship Diagram ERD . this article for: IT Students learning about database. you are having difficulties in creating ERD. you are developer that need help in database design.   let's ask ourselves Why good database design is good ? or why we need ERD?         when I started programming Eng / Eman Ahmed was my First teacher, she said you must write your code with pen, and now I told you good database design start with paper and pen or in technological terms start with ERD, ERD is the model which shows logical layout of your database. let's List some advantages of good database design : very low or no redundant data is stored.  supporting planed or unplanned queries. good design easy to maintain and modify. from first glance you can understand it. an ERD consist from Entities and relationships between them; there exist...
2 comments
Read more

Sigmoid Function

Hello all, Sigmoid Function or Logistic Function , it is used in Machine Learning problems for the purpose of learning (Neural network ) on curves. its range between ]0,1[ it is horizontal asymptotic on 0 and 1, because of its range between 0 and 1 it is used in classification using threshold for example if we take 0.5 as a threshold  values greater than 0.5 are considered Ones and  values less than or Equal  0.5 considered 0. it has the following formula :   and it shape as following : from wikipedia  so it accept differentiation over its domain unlike Step Function. differentiation of sigmoid function :  I finished the article , i hope you understand it . Thanks for your time and effort if you like it share and follow me :)
Post a Comment
Read more