i need help with this project please some one help meeee

Discussion in 'C++' started by naknak4@walla.co.il, Oct 19, 2006.

  1. Guest

    Introduction

    This assignment requires you to develop solutions to the given problem
    using several different approaches (which actually involves using three
    different STL containers). You will implement all three techniques as
    programs. In these programs, as well as solving the problem, you will
    also measure how long the program takes to run. The programs are worth
    80% of the total mark. The final 20% of the marks are awarded for a
    brief report explaining why the different algorithms performed in the
    way they did.


    The Problem - text analysis/word counting

    Computer analysis of texts, which involves making statistical
    measurements on the text, is used in a variety of situations, including
    authorship attribution (checking a piece of text by an unknown author
    against texts by others to try and work out who wrote it) and
    plagiarism detection (shudder at the thought). One of the simplest
    measures is to take a piece of text and count how often each word used
    in the text occurs. Obviously, in a large piece of text many words are
    going to be used. We are going to look at several ways of do this
    counting. We are not going to worry about analysing the data - that is
    the problem for a different program (or different part of the program)
    - we are just looking at the word counting phase. We are going to run
    various programs (six in all) which all using STL containers (some of
    the programs use the same containers but in different ways, so there
    are only three different containers used). We are going to time how
    long each algorithm takes (don't worry, the first appendix shows you
    how to do this). To enable meaningful measurements to be made, a fairly
    large piece of text is required. A file called Frank.txt has been put
    on BrightSpark for you to use. It contains the first three chapters of
    'Frankenstein' by Mary Shelley. The text has been ready prepared
    for analysis - all punctuation and capitalisation has been removed.

    Each program must read the file, and count how often each of the words
    occurs. The program then prints out a list of the words (the order of
    the words depends on which part you are doing) and how many times they
    are present in the text. There are a lot of ways that this could be
    done. Several approaches/algorithms are given below. You should measure
    the time taken to complete each algorithm. Outputting the text to the
    screen will be a slower process than doing the analysis, so you should
    take a time measurement at three points - before starting the
    analysis, after the analysis but before the output and after the
    output. You can then say how long the analysis took, how long the
    printing took and how long the overall analysis took. Remember, do not
    take a single value, but run several (say, ten times) and take the
    average.




    Note

    Most of the programs (all except 3) will need you to use a class to
    hold a word/word count pair. The following header file defines a class
    which (with the addition of the matching body) is sufficient for parts
    and 2.:






    class WordCount
    {
    private:
    string word;
    int count;
    public:
    WordCount();
    WordCount(const string& word);
    WordCount(const WordCount& wc);

    WordCount operator++(int);
    bool operator==(const WordCount& rhs);

    friend
    ostream& operator<<(ostream& os, const WordCount& rhs);
    };

    As you can see, there are two attributes, a word and the number of
    times that word occurs in the text.

    Here are some details of the operators:
    WordCount();
    This is the default constructor. You will not create objects directly
    with it, but it is required for creating some of the initial STL
    containers. It should set the count to zero and the word to an empty
    string.
    WordCount(const string& word);
    This is the constructor that you will use to create objects in your
    program). You supply the word as a parameter and set the count to 1
    (since it is used when you find a new word for the first time).
    WordCount(const WordCount& wc);
    This is a standard copy constructor which will be needed at some
    stages. You do not explicitly call it.

    Here are some details of the operators:
    operator++
    This is the increment operator (actually a post-increment) which
    increments the count field (only). Post increment operators are
    slightly strange to write, the stages are:
    a) take a copy of the object
    b) increment the count attribute
    c) return the copy of the object
    operator==
    This only compares the word fields, it returns true if they are the
    same and false if they are different. In parts 2 and 3 you will also
    need a less than comparison. It is very similar to this operator.
    operator<<
    The overloaded output operator simplifies the output stage of the
    program and should give the word followed by its count - put the
    count in brackets to make it easier to follow.

    Depending on how you write the code you may need some additional
    operators or methods. Hopefully you won't, but if you do then you may
    add them.






    Part 1

    The programs

    In this assignment, the words are output in any order, i.e. no sorting
    is done. There are three programs to write (although the second is
    only a fairly minor modification of the first):

    The first program (1)
    Use the vector container from STL. As you read in each word, check to
    see if it is already stored in the vector. If it is, increment that
    word's count. If it isn't, add the word to the end of the vector.
    It is easiest if you create a temporary word count object as soon as
    you read each word from the file and use that for the comparisons etc.

    The second program (2)
    This should be a very similar program to the first program, but should
    use a list container instead of a vector.

    The third program (3)
    This should use a hash map. The approach of the third program is
    similar to the first two, but you will not use the WordCount class
    because the hash map expects pairs of values. You will not have met
    hash maps before, so there are notes on using the hash map in appendix
    2. The hash maps work on the concept of pairs of values, one being a
    key which is used to locate the data and the other being the data
    value. If we let the word be the key and the count of that word be the
    data then the hash map will work for solving this problem. Because hash
    maps automatically work with pairs you will not use the WordCount class
    in this section.

    When you have read a word, you should use the find method to see if it
    is already in the hash map. The find method returns an iterator. If the
    iterator is set to the end of the map (i.e. is equal to wordlist.end())
    then the word is not in the map. You can use the insert method to
    insert this as a new word. Appendix 2 shows gow to use the template
    pair to do this.

    If the iterator is not set to the end then the word has been found. The
    iterator allows you to access two fields, first (the word) and second
    (the count). You can increment the count directly.

    To output the result you can declare an iterator and step through the
    hash map. You can use the first and second fields to mimic the output
    of the other programs.

    Hash maps are very efficient at insertion and retrieval, and both
    operators work in almost constant time.
    , Oct 19, 2006
    #1
    1. Advertising

  2. Sumit Rajan Guest

    wrote:
    > Introduction
    >
    > This assignment requires you to develop solutions to the given problem
    > using several different approaches (which actually involves using three
    > different STL containers). You will implement all three techniques as


    <SNIP>

    Please see:
    http://www.parashift.com/c -faq-lite/how-to-post.html#faq-5.2

    Regards,
    Sumit.
    Sumit Rajan, Oct 19, 2006
    #2
    1. Advertising

  3. SirMike Guest

    I can only say - Good luck :)

    --
    SirMike - http://www.sirmike.org

    C makes it easy to shoot yourself in the foot; C++ makes it harder, but
    when you do, it blows away your whole leg. - Bjarne Stroustrup
    SirMike, Oct 19, 2006
    #3
  4. Daniel T. Guest

    wrote:

    > class WordCount
    > {
    > private:
    > string word;
    > int count;
    > public:
    > WordCount();
    > WordCount(const string& word);
    > WordCount(const WordCount& wc);
    >
    > WordCount operator++(int);
    > bool operator==(const WordCount& rhs);
    >
    > friend
    > ostream& operator<<(ostream& os, const WordCount& rhs);
    > };


    Let's start with your WordCount class (strictly speaking, you don't need
    it for the assignment, but the teacher wants you to use it so you use
    it.)

    ----- program code -----
    #include "WordCount.h"
    #include <sstream>
    #include <iostream>
    #include <cassert>

    using namespace std;

    int main() {
    WordCount wc;
    stringstream ss;
    ss << wc;
    assert( ss.str() == "[0]" );
    cout << "Working So Far.";
    }
    ----- end -----

    Use the above and put your teachers WordCount class info in a file
    called "WordCount.h". You will also need a WordCount.cpp file:

    ----- program code -----
    #include "WordCount.h"

    // insert code as necessary

    ----- end -----

    Put code where it says "insert code as necessary" until you can make the
    program run and print out "Working So Far".

    Once you get that done, post what you have here and I'll help you with
    the next step (you can email if you like or IM me [AIM nic: ObjHead] if
    you want to go faster.)

    If you have any problems, post what you have so far and ask away!

    --
    There are two things that simply cannot be doubted, logic and perception.
    Doubt those, and you no longer have anyone to discuss your doubts with,
    nor any ability to discuss them.
    Daniel T., Oct 19, 2006
    #4
  5. Puppet_Sock Guest

    wrote:
    [assignment snipped]

    Sure naknak. Just send us your instructor's name and email, and
    we'll send himswer right away. That will save you the bother of
    having to copy it from the news group.
    Socks
    Puppet_Sock, Oct 19, 2006
    #5
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Guest
    Replies:
    0
    Views:
    678
    Guest
    Feb 25, 2004
  2. KK
    Replies:
    2
    Views:
    550
    Big Brian
    Oct 14, 2003
  3. Replies:
    5
    Views:
    757
    Andrew Thompson
    Jan 8, 2007
  4. Domino
    Replies:
    5
    Views:
    378
    dorayme
    Nov 5, 2006
  5. Replies:
    3
    Views:
    78
Loading...

Share This Page