class: center, middle, inverse, title-slide .title[ # Week 1: Welcome to the world of data science! <br>🎟🎪🎠️ ] .author[ ### Eren Bilen, Assistant Professor of Data Analytics ] .institute[ ### Dickinson College ] --- <style> .caption { text-align: center; font-size: 14px; } </style> <style> .word-under { text-decoration: none; } .def { display: none; } /* step 2: underline the word */ .remark-slide-content.step2 .word-under { text-decoration: underline; } /* step 3: keep underline and reveal the definition */ .remark-slide-content.step3 .word-under { text-decoration: underline; } .remark-slide-content.step3 .def { display: block; } </style> <!-- .caption:before { content:"Figure: "; font-weight: bold; } --> **Introduction** <blockquote>What is data science?</blockquote> &rarr;&nbsp; a field of study that uses scientific methods, processes, and systems to extract knowledge and insights from <span class="word-under">data</span>. <div class="def"><strong>Data</strong> are a collection of values that convey information, describing the quantity, quality, fact, statistics, other basic units of meaning, or simply sequences of symbols. <em>A datum</em> is an individual value in a collection of data.</div> -- class: step2 -- class: step3 -- <img src="sir.jpeg" style="position:fixed; top:250px; left:120px; width:200px; z-index:9999;"> -- `$$\\[5cm]$$` Can you think of some examples of data? -- Census, music (!), images (!), text, maps (!), among many others. --- <br><br> **Census** Every ten years, the U.S. Census Bureau carries out a full count of the population, gathering data on age, race, household structure, and housing. Why bother? -- migration, aging, population growth, family structure... -- The first census was in 1790 (!) and you can access this information!! [link](https://www.archives.gov/research/census/1790) -- But a bit earlier than this... the Babylonians are known to have conducted the first ever census in 3,800 BC. Who ordered it and why? -- Among other reasons, crop planning (you don't want your subjects to starve), tax, get a count of your labor force and military power. What does it look like?: [link](https://www.christies.com/en/lot/lot-5903835) --- <br><br> **Music (!)** What forms of information can a song contain? What do you think? -- Length, tempo, key (minor or major), loudness, lyrics (more on that in a sec), instruments. Even fancier: number of notes -- What can we do with this? -- Track an artist's career, compare artists, compare eras.. (research) -- <br> Music recommendation (🤑💸) -- <br> If you are artist: what songs "work?" (in an industrial sense) [Spotify data](https://www.kaggle.com/datasets/nelgiriyewithana/top-spotify-songs-2023) <br> [An analysis](https://www.kaggle.com/code/alankarmahajan/exploring-spotify-dataset) --- <br><br> **Images (!)** What is an image? -- A structured collection of pixels arranged in a grid. <img src="mnist.png" width="55%" style="display: block; margin: auto;" /> -- <div style="margin-top:-20px;"> Text recognition: (🤑💸) <img src="mnist2.png" width="55%" style="display: block; margin: auto;" /> </div> --- <br><br> <img src="monet.jpg" width="55%" style="display: block; margin: auto;" /> --- <br><br> <img src="picasso.png" width="85%" style="display: block; margin: auto;" /> --- <br><br> <img src="vangoghfake.jpeg" width="90%" style="margin-left:20px;" /> --- <br><br> <img src="me.jpeg" style="width:45%; margin-right:10px;"> <img src="ghibli2.png" style="width:45%;"> --- <img src="monalisa2.png" width="77%" style="margin-left:85px;" /> --- <br> <br> If you put images together, you get a ... video! Applications: [Mona Lisa rap](https://www.thestack.technology/mona-lisa-rapping-microsoft-ai-vasa-1/) [Harry Potter but in Berlin](https://www.youtube.com/watch?v=az7KfOQkMu0&ab_channel=Hoog) --- <br> <br> **Text** &rarr;&nbsp; From a sociolinguistic perspective: any symbolically encoded language. These come in many types:<br> Alphabetic – symbols (letters) to represent vowels and consonants (e.g., Latin)<br> Abjads – symbols to represent consonants, with diacritics to represent vowels (e.g., Arabic, Hebrew) <br> Syllabic – Symbol systems representing consonants plus inherent vowels (e.g., Devanagari) <br> Semanto-phonetic – Symbols that carry meaning and/or sounds (e.g., Chinese) <img src="tdm.png" width="45%" style="display: block; margin: auto;" /> --- <br> <br> **Text** Applications: Suppose you load the book Moby-Dick into your computer. &rarr;&nbsp;an 1851 epic novel by American writer Herman Melville What can you do with it? <img src="mobydick.jpg" style="position:fixed; top:220px; left:400px; width:200px; z-index:9999;"> -- <div style="width:300px;"> Word count, most frequent words, sentence length, vocabulary richness, topic modeling, sentiment analysis, character analysis, network graphs, timelines. </div> --- <br> <br> <img src="mobydick2.png" width="75%" style="display: block; margin: auto;" /> [An interactive version](https://roadtolarissa.com/whalewords) --- <div class="figure" style="text-align: center"> <img src="mobydick3.png" alt="Number of times the word &quot;whale&quot; appears on every page in Moby Dick" width="43%" /> <p class="caption">Number of times the word "whale" appears on every page in Moby Dick</p> </div> --- <br> <br> **Text** More applications: authorship classification, spam filter (🤑💸) <br> <div style="display:flex; align-items:flex-start;"> <figure style="width:45%; margin-right:10px;"> <img src="voynich.jpg" style="width:100%; display:block;"> <figcaption style="text-align:center; font-size:14px; margin-top:6px;"> Voynich manuscript </figcaption> </figure> <figure style="width:50%; position:relative; top:20px;"> <img src="spam.png" style="width:100%; display:block;"> <figcaption style="text-align:center; font-size:14px; margin-top:6px;"> Spam detection </figcaption> </figure> </div> --- <br><br> <img src="aiarticles.png" width="80%" style="display: block; margin: auto;" /> --- <br> <br> **Maps (!)** &rarr;&nbsp; contain information on where borders are, geographical information, demographic information.. among others. <br><br> <img src="maps.png" width="60%" style="display: block; margin: auto;" /> --- <br><br> <img src="pa1.jpg" width="90%" style="display: block; margin: auto;" /> --- <br><br> <img src="pa2.jpg" width="90%" style="display: block; margin: auto;" /> --- <br><br> <img src="pa3.png" width="90%" style="display: block; margin: auto;" /> -- [Interactive map: PA house of representative districts](https://www.redistricting.state.pa.us/Maps) --- <br><br> <img src="mcdonalds.png" width="90%" style="display: block; margin: auto;" /> -- [Interactive map: All McDonalds locations in the U.S.](https://www.arcgis.com/home/item.html?id=2a6d0ecb7e5645afb49d8f38a2d2b87b) --- <br><br><br> <img src="thinking.png" width="55%" style="display: block; margin: auto;" /> <div style="text-align: center;"> Questions..? <br><br> </div>