You are going to write a complete C program which implements the following functionality:
- Your program will read the following files:
- Each file contains text in a specific language. All files contain only english lowercase characters and whitespace. Text files will include the following characters:
‘a’ ‘b’ ‘c’ ‘d’ ‘e’ ‘f’ ‘g’ ‘h’ ‘i’ ‘j’ ‘k’ ‘l’ ‘m’ ‘n’ ‘o’ ‘p’ ‘q’ ‘r’ ‘s’ ‘t’ ‘u’ ‘v’ ‘w’ ‘x’ ‘y’ ‘z’ ‘ ‘
- Your program will evaluate the dissimilarity scores of language pairs:
- First of all, calculate bi-gram frequencies for each language. A bi-gram is defined as follows: For a given sequence, each unique pairing of successive letters is a bi-gram. For example: for the word ” adana ” bi-grams are defined to be ” a”, “ad”, “da”, “an”, “na”, “a “. Beware: If there is a space before or after a character you will still be dealing with bi-grams. Each bi-gram has exactly two elements which are either characters or space. In order to calculate the frequency of a particular bi-gram(lets say bi-gram “ad”) you have to count all the bi-grams in a given text and for this bi-gram calculate the ratio (# of “ad”)/(total # of all bi-grams)
- Given all the frequencies, dissimilarity score is calculated as follows:
dissimilarity(languagea, languageb) = |fai − fbi| (1)
- Here fai represents the frequency of ith bi-gram for the language a. If cia is the count of ith bi-gram in languagea, then;
fi = ci /( cj ) (2)
a a a
- After evaluating dissimilarities, your program will print all the dissimilarity values. Print:
- text files can include multiple concatenating whitespace. For example:
Here we are using a user defined recursive
- Two adjacent whitespace do not create a bi-gram.
- Input files can be multi-line text files.
- There isn’t any limit on the size of input files. Your program should work regardless of the size of the input.
- Bi-gram types do not depend on the input file. There are finite number of possibilities. Given all the lowercase english characters and a space, you can generate all the possible bi-grams.
- Do not try to store all the content of the file in the memory. Counting is possible without storing all of the text.
- You don’t need to parse words.
A complete C program <assignment_5_name_id.c> which can be compiled using the following command:
gcc -std=c99 assignment_5_name_id.c -o assignment_5_name_id
If your program requires additional compile and link options, state that require-ment at beginning of your source code as a comment.
- Read and apply “Assignment Submission Rules and Other Related Infor-mation” document which available on the class e-learning system.
- You may or may not get partial credit depending on how you structured or documented your code.