Howto

FoodReader.py

Function: get_data( food ) : The function must read data about food and raters and return the information in a list and dictionary. (see details)

The food argument must be a txt file of food and ratings (supplied with the snarf or get here):

rater

      food rating, food rating, food rating

      rater 

      food rating, food rating

      rater 

      food rating, food rating, food rating, food rating

      ....

For each rater, the rater is on one line and a list of one or more restaurants and integer ratings appear on the next line separated by commas, if there is more than one.

Return values

The call get_data( food) must return two things:

itlist a list of items being rated: [ "restaurant", "restaurant", ... ]

rdict a dictionary of raters and their ratings of the restaurants:

{"ratername" : [integer ratings for each restaurant],
 "ratername" : [integer ratings for each restaurant], ...
}

BookReader.py

Function: get_data( books, ratings ) : The function must read data about books and raters and return the information in a list and dictionary. (see details)

The books argument must be a txt file of books (supplied with the snarf or get here):

author,bookTitle

      author,bookTitle

      ....

The ratings argument must be a txt file of raters (supplied with the snarf or get here):
ratername
bookRatings
ratername

bookRatings

      ....

Return values

The call get_data( books, ratings ) must return two things:

itlist a list of items being rated: [ "author,bookTitle", "author,bookTitle", ... ]

rdict a dictionary of raters and their ratings of the books:

{"ratername" : [integer ratings for each movie],
 "ratername" : [integer ratings for each movie], ...
}

MovieReader.py

Function: get_data( movieratings )
(see details)

The movieratings argument must be a txt file of movies (supplied with the snarf or get here):
ratername title rating
ratername title rating

ratername title rating

      ....

Return values

The get_data( movieratings ) must return two things:

itlist a list of items being rated: [ "title ", "title ", "title ", ... ]
(see details and examples here)

rdict a dictionary of raters and their ratings of the movies:

{"ratername" : [integer ratings for each movie],
 "ratername" : [integer ratings for each movie], ...
}

Note that all functions get_data return the same types: a list of strings (each item being rated) and a dictionary in which keys are raters and values are lists of integer ratings by each rater.

Recommender.py

In the module Recommender.py you must write the three functions shown in the details pages: averages, similarities, and recommended. You must also (either in Recommender.py or another module you write that imports and calls Recommender) obtain the list and dictionary from either a book or movie module's get_data function, and then report on at three different ratings:

For all three of the following, make sure your output is easy to understand. Print a statement before each of the items below to identify what you are printing.

Print a sorted list of every item being rated and its average rating, sorted by rating with the highest rating printed first (use averages).
For a specific user (Owen or someone for whom ratings are given) print a list of all other raters sorted by "closeness" to the specific user, with the closest rater listed first. (use similarities).
Print a list of 20 items for the specific user that are "most" recommended as determined by a weighted rating. (use recommended).

(See details)

README/Analysis

You should submit a README file as usual. You must also submit either a text file or a .doc or .pdf that is your analysis of what you've done. The analysis should include the following:

A description of how you determined that the three functions you wrote in Recommender.py are correct. What did you do to verify the results, how did you debug the code, and so on.
A description of results generated by calling recommended with different values of n for one of the data sets to determine what items are recommended.

Suggested Order to work on:

Write the FoodReader.py module. It's data file is the smallest, so it should be easier to debug.
Write the BookReader.py module or the MovieReader.py module, but not both.
Write the three functions in Recommender.py in order from simplest to most complex, with average first, then similarities, then recommended.
If doing the extra credit, then attempt it.

Howto

FoodReader.py

Return values

BookReader.py

Return values

MovieReader.py

Return values

Recommender.py

README/Analysis

Suggested Order to work on:

Next 'details'