reiver-artificial-intelligence/levels-of-measurement/index.xhtml

274 lines
6.9 KiB
HTML
Raw Blame History

This file contains ambiguous Unicode characters!

This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.

<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" lang="en">
<head>
<meta charset="utf-8" />
<title>Levels of Measurement</title>
</head>
<body>
<article>
<hgroup>
<h1>Levels of Measurement</h1>
<p><small>(<a href="../">Artificial Intelligence</a>)</small></p>
</hgroup>
<section>
<address class="h-card">
by
<a rel="author" class="u-url" href="http://changelog.ca/"><span class="p-given-name">Charles</span> <span class="p-additional-name">Iliya</span> <span class="p-family-name">Krempeaux</span></a>
</address>
</section>
<section>
<p>
In <strong>artificial-intelligence</strong>, you can end up dealing with all sorts of data.
For example —
</p>
<ul>
<li>maybe you are working with <em>photos</em> of people's faces;</li>
<li>maybe you are creating a list of <em>city names</em> of cities in the Near East;</li>
<li>maybe you have a list of <em>blood types</em> for all the patients that have visited the hospital;</li>
<li>maybe you are working with hospital patient data about how severe their pain was — ex: "<em>no pain</em>", "<em>mild pain</em>" , "<em>moderate pain</em>", "<em>it hurts bad</em>", and "<em><abbr title="oh my fucking! god">OMF!G</abbr>, this is the worst pain I have ever felt!</em>";</li>
<li>maybe you are looking at student <em>test scores</em>, graded oon a curve;</li>
<li>maybe you are recording the <em>birth order</em> (i.e., 1st child, 2nd child, 3rd child, etc) of all the people who work in a company;</li>
<li>maybe you are have <em>temperature<.em> records in °C (degree celsius);</li>
<li>maybe you are looking at the year every customer of a company was born;</li>
<li>maybe you are recording the <em>heights</em> of all the children in an elementary school;</li>
<li>maybe you are collecting the prices of houses in the city of Isfahan;</li>
<li>maybe you are are looking the number of photo each person takes every day on their mobile-phone;</li>
<li>etc etc etc.</li>
</ul>
</section>
<section>
<h2>Some Differences</h2>
<p>
Some of these data are very different than each other.
</p>
<p>
For example — if the price of one house is $1,000,000 and the price of another house is $500,000, then I can say the price of the first house is 2× (two times) bigger than the first.
(Since $1,000,000 = 2 × $500,000.)
</p>
<p>
But it is nonsense to say — <em>2 × type O+ blood</em>.
There is no concept of multiplying blood types.
</p>
<p>
The same is true for city names.
It is nonsense to say — <em>5 × the city of Vancouver</em>.
</p>
</section>
<section>
<h2>Types of Data</h2>
<p>
The <strong>levels of measurement</strong> are a way of understanding these <em>different types of data</em>.
</p>
<p>
The <strong>levels of measurement</strong> are a way of categorizing a <ziba-link>data</ziba-link> as one of four different categories:
</p>
<ul>
<li><ziba-link>nominal data</ziba-link>,</li>
<li><ziba-link>ordinal data</ziba-link>,</li>
<li><ziba-link>interval data</ziba-link>, and</li>
<li><ziba-link>ratio data</ziba-link>.</li>
</ul>
</section>
<section>
<h2>Ratio Data</h2>
</section>
<section>
<h2>Interval Data</h2>
</section>
<section>
<h2>Ordinal Data</h2>
</section>
<section>
<h2>Nominal Data</h2>
<section>
<p>
An example of <strong>nominal data</strong> is <em>blood types</em>:
</p>
<ul>
<li>type A+</li>
<li>type A-</li>
<li>type B+</li>
<li>type B-</li>
<li>type AB+</li>
<li>type AB-</li>
<li>type O+</li>
<li>type O-</li>
</ul>
</section>
<section>
<p>
Another example of <strong>nominal data</strong> is <em>sex</em>:
</p>
<ul>
<li>female</li>
<li>male</li>
</ul>
</section>
<section>
<p>
Another example of <strong>nominal data</strong> is <em>family names</em>:
</p>
<ul>
<li>Alves</li>
<li>Beg</li>
<li>Chen</li>
<li>Cho</li>
<li>Choi</li>
<li>da Silva</li>
<li>Dickson</li>
<li>dos Santos</li>
<li>Esfahani</li>
<li>Fernández</li>
<li>Ferreira</li>
<li>García</li>
<li>Jung</li>
<li>Kang</li>
<li>Kerr</li>
<li>Kim</li>
<li>Krempeaux</li>
<li>Li</li>
<li>Liu</li>
<li>Martin</li>
<li>Müller</li>
<li>Pahlavi</li>
<li>Park</li>
<li>Parsi</li>
<li>Pereira</li>
<li>Rodríguez</li>
<li>Safavi</li>
<li>Sasani</li>
<li>Wang</li>
<li>Yun</li>
<li>Zhang</li>
<li><em>etc</em></li>
</ul>
</section>
<section>
<p>
Another example of <strong>nominal data</strong> is <em>hair color</em>:
</p>
<ul>
<li>brown</li>
<li>black</li>
<li>blond</li>
<li>gray</li>
<li>red</li>
<li><em>etc</em></li>
</ul>
</section>
<section>
<p>
Another example of <strong>nominal data</strong> is <em>cities</em>:
</p>
<ul>
<li>Beijing</li>
<li>Bangalore</li>
<li>Bangkok</li>
<li>Bogotá</li>
<li>Buenos Aires</li>
<li>Cairo</li>
<li>Chennai</li>
<li>Chicago</li>
<li>Chongqing</li>
<li>Dallas</li>
<li>Delhi</li>
<li>Dhaka</li>
<li>Guangzhou</li>
<li>Hyderabad</li>
<li>Kinshasa</li>
<li>Kolkata</li>
<li>Isfahan</li>
<li>Istanbul</li>
<li>Jakarta</li>
<li>Karachi</li>
<li>Lagos</li>
<li>Lahore</li>
<li>Lima</li>
<li>London</li>
<li>Los Angeles</li>
<li>Moscow</li>
<li>Nagoya</li>
<li>New York City</li>
<li>Osaka</li>
<li>Manila</li>
<li>Mexico City</li>
<li>Mumbai</li>
<li>Paris</li>
<li>Rio de Janeiro</li>
<li>São Paulo</li>
<li>Seoul</li>
<li>Shanghai</li>
<li>Shenzhen</li>
<li>Tehran</li>
<li>Tianjin</li>
<li>Tokyo</li>
<li>Toronto</li>
<li>Vancouver</li>
<li><em>etc</em></li>
</ul>
</section>
</section>
<section>
<h2>Operations</h2>
<p>
One way of understanding the <strong>levels of measurement</strong> is — what type of <strong>operations</strong> are meaningful and valid for that type of data‽
</p>
<p>
Here is a table that summarizes what <strong>operations</strong> are and are <em>not</em >meaningful and valid for each <strong>levels of measurement</strong> category:
</p>
<table>
<thead>
<tr>
<th></th>
<th>nominal</th>
<th>ordinal</th>
<th>interval</th>
<th>ratio</th>
</tr>
</thead>
<tbody>
<tr>
<td>can × and ÷ ?</td>
<td>no</td>
<td>no</td>
<td>no</td>
<td>YES</td>
</tr>
<tr>
<td>can + and - ?</td>
<td>no</td>
<td>no</td>
<td>YES</td>
<td>YES</td>
</tr>
<tr>
<td>can &lt; and &gt; ?</td>
<td>no</td>
<td>YES</td>
<td>YES</td>
<td>YES</td>
</tr>
<tr>
<td>can = and ≠ ?</td>
<td>YES</td>
<td>YES</td>
<td>YES</td>
<td>YES</td>
</tr>
</tbody>
</table>
</section>
</article>
</body>
</html>