2.2.2 Computational Methods Flashcards
What are the 6 key principles of computational methods?
- Decompositon
- Pattern recognition
- Abstraction
- Algorithms
- Calculations
- Storage
Define ‘backtracking’
A method of revisiting states in a problem to examine alternative routes
How does backtracking work?
- Often using recursion, the program will move from one state of the problem to another
- A decision is made from examining possible options
- Previous states will be returned to if one path is eventually rejected
Define ‘data mining’
The process of digging through data sets to find connections and predict trends
How does data mining work?
Unnoticed patterns are revealed and underlying relationships discovered within the data.
This allows informed decisions to be made
Define ‘heuristics’
A heuristic approach applies algorithms or methodologies which are not guaranteed to be optimal.
The objective is a solution in a reasonable time frame
What is the benefit of using heuristics?
Allows problems to be solved in much shorter time frames
Define ‘performance modelling’
The process of simulating different users and systems on a computer with mathematical approximations
How does performance modelling work?
Rather than performing actual tests, these simulations are cheaper and quicker to run and analyse
Describe ‘pipelining’
Splitting tasks into smaller parts and overlapping the processing of the tasks
How does pipelining work?
The Fetch, Decode, and Execute stages of several processes are processed at the same time
Explain ‘visualisation to solve problems’
Using graphical techniques to illustrate a problem, and how to approach the solution
Describe how visualising to solve the problem is used
These techniques can be used to explain tasks, model processes, present information, and offer alternative views to humans, which can then better solve problems
