[SOLVED] Implementing 8-puzzle search in java with full source code | download now

By Anonymous -

The 8-puzzle problem

In this project you will implement a solution to the 8-puzzle by state-space search, using the search engine described in the lectures, and experiment with search strategies.

2. What you must do

2.1 Implementing 8-puzzle search

In the search2 folder in the Java download on the COM1005/2007 MOLE site is the code for the simple search engine and for Jugs problems. Note that you may have to recompile the code for your version of Java.

  • Following the instructions in section 2.9 of the lecture notes, write classes to implement a state-space search for 8-puzzle problems.
  • You should not need to change the code for the search engine except perhaps to control how much it prints as a search proceeds, and to stop the search after a given number of iterations (in an 8-puzzle the search tree can become large).
  • Test your implementation with breadth-first searches for the following initial con- figurations and the same goal state as above:


2.2 Experimenting with Search Strategies

The search engine in the search4 folder implements the following search strategies:

  • Breadth-first
  • Depth-first
  • Branch-and-Bound
  • A*

In this case we won’t consider Depth-first and Branch-and-Bound is the same as Breadth- first because we have uniform costs. We’ll compare Breadth-first with 2 variants of A*.


  • Adapt your 8-puzzle classes for search4. This means that your 8-puzzle state must now include a local cost g (always 1 for the 8-puzzle), and an estimated remaining cost estRemCost, which is used by A* and must be an underestimate of the true cost.
  • Code two alternative methods in your 8-puzzle state class for computing estRemCost, assuming that the target pattern is always the same, as above.
    • Hamming distance, which is the number of tiles out of place.
    • Manhattan distance, which is the sum of the moves each tile needs to make before it is in its correct position.

The experiment is to compare the efficiency of breadth-first, A*(Hamming) and A* (Manhattan) over a number of puzzles. You are testing the hypothesis that A* is more efficient than breath-first, and the efficiency gain is greater the more difficult the problem and the closer the estimates are to the true cost.

Get your project now 

Buy now

Comments

  1. safronmachak3 March 2022 at 04:59

    Casino Slot Games - Mapyro
    Looking for 강원도 출장샵 Casino Slot 부산광역 출장마사지 Games near you? Mapyro provides you 충주 출장안마 with 계룡 출장안마 the closest casino to your location with over 광명 출장안마 60000 FREE slot What is the Best Casino Slot Machine to Play at?What is the Best Online Slot Machines to Play at?

    ReplyDelete

Post a Comment

Popular posts from this blog

[SOLVED] Tape for a Turing Machine using Doubly-linked List in Java with full source code

By Anonymous -
Tape for a Turing Machine (Doubly-linked List) Short exercise in which you will code part of an existing project. In this case, you will be working with a doubly-linked list. To make this topic more interesting, the list that you work on will be part of a Turing Machine. A Turing Machine is a kind of very simple, abstract computing device that was introduced in the 1930's by Alan Turing as a way of studying computation and its limitations. The classes that you need for the lab are in a package named turing. You will define a new class named Tape in this package. You can find the Java source in the code directory. You should copy this directory into an Eclipse project. There will be errors, since the existing classes depend on the Tape class, which you have not yet written. Don't forget the Javadoc comments. A Class for Turing Machine Tapes A Turing machine works on a "tape" that is used for both input and output. The tape is made up of little squares called cell
Read more

[SOLVED] Branch Coverage, Statement Coverage and Path Coverage | Java Unit Testing

By Anonymous -
Create a directory called “TestClassValue” Hereafter, we call this directory <dir>. Create a Java class com.work.sample.FlawedClass in directory <dir>/src. (The actual path will obviously reflect the package structure.) 100% Branch Coverage & 100% Statement Coverage Add to the class a method called flawedMethodl that contains a division by zero fault such that it is possible to create a test suite that achieves 100% branch coverage and does not reveal the fault every test suite that achieves 100% statement coverage reveals the fault. Conversely, if you were able to create the method, then create two JUnit test classes com.work.sample.FlawedClassTestBCl and com.work.sample.FlawedClassTestSCl for class FlawedClass as follows: FlawedClassTestBCl should achieve 100% branch coverage of flawedMethodl and not reveal the fault therein. FlawedClassTestSCl should achieve 100% statement coverage of flawedMethodl and reveal the fault therein. Both classes should be
Read more

[SOLVED] Buggy Search And Sort Buy Reporting with Java source code

By Anonymous -
Report The Bug From The Following Source Code: import java.util.Arrays; /** * This class looks like it's meant to provide a few public static methods for * searching and sorting arrays. It also has a main method that tests the * searching and sorting methods. *  * TODO: The search and sort methods in this class contain bugs that can cause * incorrect output or infinite loops. Use the Eclipse debugger to find the bugs * and fix them */ public class BuggySearchAndSort { public static void main(String[] args) { int[] A = new int[10]; // Create an array and fill it with random ints. for (int i = 0; i < 10; i++) { A[i] = 1 + (int) (10 * Math.random()); } int[] B = A.clone(); // Make copies of the array. int[] C = A.clone(); int[] D = A.clone(); System.out.print("The array is:"); printArray(A); if (contains(A, 5)) { System.out.println("This array DOES contain 5."); } else { System.out.println("This array DOES NOT contain 5."
Read more