Data ECE242: Structures and Algorithms PROGRAMMING PROJECT 4
BINARY SEARCH TREES AND HASH TABLES
NOTE : WE WILL BE USING THE SPECIFICATIONS AS A BASIS FOR EVALUATION. SO PLEASE CONFORM TO THE SPECIFICATIONS TO RECEIVE FULL CREDIT. Introduction and Problem Abstract: The ECS Office is planning to re-structure its student database and after a lot of planning, it has been finally decided that it is to be implemented as a combination of a chained hash table (e.g. slide 16 of lecture 26) and a number of binary search trees which use linked structures (e.g. see code for lecture 24). You have been asked to develop a prototype for the new database implementation. At a high level, your code will work as follows for adding a new course to a student's academic record in the database : 1. A student UMass ID is used to generate a hash value. This value is used to locate one specific index in the array of student nodes. 2. The student node at that index is searched for the relevant details. If present, it is updated. If not, it is added at that index. 3. Each node contains a reference to the BST. The BST is ordered by the names of the student's courses. Now consider the specific steps needed to complete this project. Please go over the problem specification carefully before you start coding. Figure 1 is an example for reference before we go into the details. An example : HASH TABLE 0 1 2 3 . . . . . . 99
Student ID = 23463299 Name = Fred nextNode = DEF Reference to BST Student ID = 23463399 Name = Bill NextNode = null Reference to BST Algo Analysis B Field Theory A Data Structures A
Student ID = 23463280 Name = Abhi NextNode = null Reference to BST
Figure 1
It should be noted that all student nodes will have their respective BSTs. The BST for only one node is shown here as an example. Problem Specifications: This section gives you the low level design description of each class and what each one contains. Part 1: Part 1A : This specification is for class studentNode. Every object of this class denotes one single student record that holds the UMASS student ID, name and course list. Define a class studentNode, with the following data members. 1. studentID This is of type long and will hold the 8 digit UMASS Student ID. 2. studentName This is of type String and will hold the name of the student. 3. BSTree object This is the reference to the Binary Search Tree that will hold the list of the students courses and their respective grades. 4. studentNode nextNode This is of type studentNode. It holds the reference to the next node whose studentID has also been hashed to the same index in the hash table. All the data members listed above are PRIVATE data members. An array of studentNodes will act as your hash table that is detailed in the next section. You can define a constructor to initialize these data members to their default values such as 0 and null if you want to do so. Part 1B : This specification is for class hashTable. This class has methods that support insertion / searching / modification of student records in the hash table. Define a class hashTable with the following data members and methods : 1. hashTableArray A private array of 100 objects of type studentNode. This array serves as your hash table to store student records. 2. addToTable() - This is a member method of the class that is used to add a studentNode (student record) to the hash table. It takes as parameters the hash value (generated from the hash function), student ID, student name, course name and course grade. Use the hash value to index into the hash table. Check all existing nodes at that index to see if the record associated with the student ID exists. If the record does not exist, insert a new record at that index. Print a message to the user "The record did not exist and was inserted at index <index #>".
If any other record already exists at the same index, a collision has occurred. We will use chaining to resolve the collision. When a collision occurs, the nextNode data member of the node already present at that index will be updated to point to the new node. You can refer to the example figure in the beginning for clarity. Print a message to the user "The record did not exist and was chained at index <index #>".
If the student record already exists at the given index, traverse the BST tree associated with the student record and check whether the input course is present in the BST or not. If the course is present, update the BST node with the input grade. Print a message that says "Record found at index <index #> and course information was updated." If the course is not present, insert a new node into the BST. Print a message that says "Record found at index <index #> and course information was inserted.". You may refer to the code available in Lecture 24 to implement this feature. Please refer to the flowchart in figure 2 for better understanding. 3. printStudent() : This method is used to search the hash table for student details and print them on the screen. It takes the hash value and studentID as inputs. Index the array of student nodes using the hash value and then retrieve the information present at that index, including the course information (hint : you need to traverse the BST again here) and print it out on the screen. In case multiple nodes exist at the same index (hash collisions), use the studentID to retrieve the correct record. If no record is found, print a message "Invalid Student ID. No record found for student."
Flowchart for addToTable() :
Hash value from hash function.
NO
Is studentNode found at index = hash value?
YES
NO
Is space at index empty or does it contain another node?
YES
NO
Is input course found in BST associated with student node?
YES
Create new node for student record being added and update existing node's nextNode to refer to the newly created node. Print message to user.
Create new node for student record being added and insert it at that index. Print Message to user as specified above.
Update node in BST that contains this course with grade that is being Print passed. message to user as specified above.
Insert new node into BST with course name and grade.
Figure 2
Part 1C : This specification is for the class BSTNode. This class defines the contents of each node of the Binary Search Tree that is going to store the course information. Define class BSTNode which will contain the following data members : String courseName char courseGrade BSTNode leftChild BSTNode rightChild This is just a small modification to the nodes that were presented during the lecture where they stored the ID, Name and Score. Again, all these data members are supposed to be private data members. Part 1D : Define class BSTree and add methods to this class that will facilitate addition into the BST and traversal of the BST. You can use the code given to you in Lecture 24 for addition and traversal. Note that you DO NOT need to include the remove() method. Part 2 : This specification is for the driver class where you will write the main method and the hash function. Define a class driverHashTable and write a private method called generateHashValue. This method takes as input an 8-digit UMASS student ID and generates a hash value between 0-99, both inclusive. It CANNOT USE ONLY modulo arithmetic. It needs to have some other operation working with modulo arithmetic to produce the hash value. Better hash functions will result in lesser collisions. In the main method, open the input file given to you. Read each line in the file, parse the student details and add them to the hash table. Next, search for a valid student ID. Verify your results. Also, search for an invalid student ID and verify your results. POINTS TO KEEP IN MIND : Most importantly, the hash value that you generate from the hash function is used to index into the array where a reference to the student node is stored (data pertaining to the student). So your hash values should map to your array indices with respect to this project. 1. This is a project that is a bit more complex than the others and involves quite a bit of coding. Please do not retain it for the last weekend before the submission date. 2. Please note that there will be a minimum of 5 classes in this project, class hashNode, class HashTable, class BSTNode, class BSTree (from the lecture) and the driver class. You are free to add more if you want to. 3. Please conform to the standards to make evaluation easier. You can introduce new functionality if you want, but please retain the class names and members as they are. 4. Please clarify any questions in the problem specifications well before hand and do not keep them for the last minute. Every effort has been made to make this a totally unambiguous specification.
5. Please refer to and post any questions on the Project 4 section of the forum.
APPENDIX A : Sample input calls to the methods and the outputs that they are expected to produce are provided below. You can relate to figure 1, array index 99 where first the record for Fred is inserted (23463299) and then the record for Bill is chained (23463399) and finally, Fred's record is updated again with a new grade. 1. Case 1 : This is the case where the UMASS Student ID hashes to an index and there is no record present at that index. addToTable(99, 23463299, "Fred", "Computer Networks", 'A'); should produce a message that says "Record did not exist and was inserted at index 99"; 2. Case 2 : This is the case where the UMASS Student ID hashes to an index which already has a record existing and the new one needs to be chained there. addToTable(99, 23463399, "Bill", "Data Structures", 'B'); should produce a message that says "Record di...