Meet the datasets: marks, bills and words

A dataset is a collection of related items. In this course, each item is drawn as a card. Each card has named fields. Different cards follow the same format, while the field values change.

Marks cards #

A marks card records sequence number, name, group/gender label, date of birth, town/city, maths marks, physics marks, chemistry marks and their total. The sequence number identifies the card. Two students can share a name, so a name alone is not a reliable unique ID.

Shopping bills #

A simplified bill records a sequence number, store name, customer name, purchased items and total bill value. Each item row records item name, category, quantity, unit price and cost. The course omits real-bill details such as tax when they are not needed for the task.

One bill is one outer record, even though it contains several item rows. “Number of bills”, “number of item rows” and “total purchased quantity” are different questions.

Word cards #

The lecture takes words from a paragraph and creates one card per occurrence. A word card stores its position, word text, part-of-speech category and letter count. Repeated words get separate cards because they occur at different positions. Punctuation stays attached to the word in the lecture dataset.

For the teaching word “bright”, the category is Adjective and letter count is 6. For “open”, the part of speech depends on the sentence: “open the door” uses a verb, while “the door is open” uses an adjective.

The questions determine the fields you need #

Source: Lecture 1.1 — Introduction to datasets. Small teaching datasets and diagrams are prepared for these notes; lecture and tutorial totals are identified explicitly.