OPTIMIZATION ON SECURITY SYSTEM WITH GAME THEORETICAL APPROACH
A good website security system is needed to protect the information contained on the website. In the field of information technology, the existence of a bug can disrupt the security system and can be used by hacker to break-ins the website which is protected by the web master. Game theory is one of...
Saved in:
Main Author: | |
---|---|
Format: | Final Project |
Language: | Indonesia |
Online Access: | https://digilib.itb.ac.id/gdl/view/27085 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Institut Teknologi Bandung |
Language: | Indonesia |
Summary: | A good website security system is needed to protect the information contained on the website. In the field of information technology, the existence of a bug can disrupt the security system and can be used by hacker to break-ins the website which is protected by the web master. Game theory is one of the approaches that can be used to study the interaction between two players in the case of website security system to get optimal results for each player. By using the game theory approach, mathematical results will be obtained for each player as a decision that can be confirmed through the idea of game theory. The case of website security system will be assumed to be a stackelberg security games model, which is a non-cooperative sequential game with the web master as a leader who will choose his strategy first and continued by the hacker as a follower who will choose the strategy according to the last condition of the game. Each chosen strategy will provide a payoff for each player and there will be a reward which is calculated from the expectations of each strategy selected in a game. The maximizing reward problem is considered as an optimization problem that will be solved by using mixed integer linear programming method. The results shown that the reward of each game depends on how the payoff matrix is constructed. The calculation with a non-uniform probablity assumption gives a more optimal result rather than the uniform probability assumption in each of player strategies. It is also shown that the substitution of follower type from hacker to lamer will still give an optimal result for the leader. |
---|