Sunday, 8 October 2017

Data Mining, Artificial Intelligence, Machine Learning- Get towards Learning

The terms Data Mining, Artificial Intelligence (AI) and Machine Learning are among the most widely searched topics on the internet. Their usage is widespread and used across various industries ranging from finance to healthcare. But, there is always some confusion on how are these terms related to each other conceptually. Let’s take a deep dive into these concepts one by one starting with Data Mining.

Data Mining

Conversion of raw Data into Useful information so that it can help the end-user in taking the decisions.
Lets’ discuss the cycle followed by Data Mining.
  1. Data Requirement: For any analysis of problem, related input is required which present in the form of raw data. This data may be numerical or categorical.
  2. Data Collection: The raw data can be take out from different sources like throgh online engines, stored directories etc.
  3. Data Processing: Before analysis of data, it is important to organised data or placed data in some stored format like in excels, database, or using some software.
  4. Data Cleaning: Now, this raw data contains such type of data which is not useful for the analysis and make the analysis more complex like, duplicates, unmatched, inaccurate records, fake data. These should be cleared before going forward.
  5. Data Analysis: Now, the data is ready for analysing and understanding. There are different problem solving, statistics, mathematical algorithms; programming languages which are applied to actually understanding the information present in this data. 



The applications of data mining are very vast. Artificial Intelligence and Machine learning are also one of the parts of Data Mining. It is also very much useful in Business for analysing Customer behaviour for different products, analysing feedback of the customers, searching online etc. We have one very common application of Data Mining which is using all over the world is Google Search.

Artificial Intelligence
  • Smartness
  • Intelligence

Both these behavior are provided artificially to the machines or computers.

How we can achieve AI?? Big challenge!!

Several Technologies are already created and still in research so that it become easy for machine to behave intelligently. Some of the common technologies which make the AI the most searched and hot topic in today’s world are:

  • Machine learning:  Implementation of different algorithms on large dataset through continuous learning in order to make prediction.
  • Natural Language Processing: It helps the machine to understand the human behaviour and language.
  • Perception: Machine takes the input in form of signals or we can say using different sensors. And the most common example in this are face, speech, Handwriting Recognition.
  • Robotics: provided intelligence to Robot. Combination of Computers, Mechanical, Electronics. Intelligent robots adapt to new environment and learn from their mistakes.
  • Reasoning, Problem Solving: AI increases the execution speed of complex programs by solving the complex problems. 
  • Planning and Scheduling

The AI is used in every sector whether it is Healthcare for Medical analysis and diagnosis, Automation or Finance. It is also used for entertainment purpose. Thus, we can provide intelligence to AI by different ways like Learning, Perception, problem solving, Reasoning (taking, judgement, decisions, predictions).

There are various Algorithms which are mostly used in AI are: Neural network, Fuzzy Logic, Heuristic Approach, Genetic Algorithm, different algorithms of PERT/CPM.

Machine Learning

Machine Learning is the successful approach of Artificial Intelligence or it is one of the methods of Data Analysis in which computers will able to learn and predictions can be done through the data.

The Core principle behind Machine learning is that machines takes large data sets and apply algorithms so that machines learns and make predictions with least error. It means prediction using ML improves with experience.

The algorithms are used to find the pattern from the given information of data in Data Mining, but, Machine learning has the extra ability to change the program behaviour after learning thousands of times and so it makes good prediction with least error.

The common applications where machine learning used are presented in the figure.
In Image recognition, the machine can predict from the data set of pictures and after many repetitions, the machine will able to learn which type of pixels are using and can make good predictions from the images.

Thanks all for reading this blog!!

You can find more information on these topic from the below reference which help me to take knowledge before writing this blog.



on

2 comments:

Subscribe to: Post Comments (Atom)