saimasharleen / book-data-generator Goto Github PK

This project is a part of making a system that suggests queries. Here, we focused mostly on assembling the dataset that will be utilized for the project.

Here's why we created four types of dataset here:

• A relational table populated with tuples. The table is expressed as a CSV file, where each row is a tuple, and the first row contains the names of the fields (attributes). You can assume that there are no NULL values. All the fields of all the tuples have a value.
• A user set. A user set was then built. The collection of users consists of merely an array of ids. This data is provided in a file format, with one user ID per line.
• A query set, which is a set of queries that have been posed in the past. Each query has a unique query id and its definition (i.e., the set of conditions). The query set is expressed as a CSV file where each line contains the conditions attribute=value separated by comma. The first element of each row is a query id. For example, a row in the query set is: Q1821, author=John, genre=fantasy
• A utility matrix User-Query that has for certain combinations of user and query, the User-Query utility matrix contains, for each combination of user and query, a value between 1 and 100 that represents the user's level of satisfaction with the result set for that query. This is also offered as a CSV file, with the first row including a list of query ids separated by commas. Each alternate row begins with a userId, then a sequence of fields separated by commas (as many as the number of query names of the first row). These fields may be left empty or populated with numbers between 1 and 100. We created this file with 20% missing value.

• [Python][ https://www.python.org/ ][ Python-url]

(back to top)