34:["$","title",null,{"children":"Exploratory Emoji Sentiment Analysis of Tweets in 2015 | Xumin Zhu | Observable"}] 3c:T7c6,{ // create an interval selection over an x-axis encoding const brush = vl.selectInterval().encodings('x'); // determine opacity based on brush const opacity = vl.opacity().if(brush, vl.value(0.9)).value(0.1); // an overview histogram of used emojis in different categories // add the interval brush to select cars over time const overview = vl.markBar() .encode( vl.x().fieldN('Unicode block') // extract emonji catefories, treat as norminal .sort({field:'y', order: "ascending" }) .scale({type:'sqrt'}) // subbdivide the full default chart width .axis({title: 'Emoji Category', labelAngle: -45}), // no title, no label angle vl.y().count().title('Occurrences') // counts, and axis title .scale({type: 'log'}), opacity // modulate bar opacity based on the brush selection ) .select(brush) // add interval brush selection to the chart .height(150) // set chart height to 120 pixels .width(700) // set chart width to 1200 pixels .title('Occurrences of the most popular emojis categories (use brush to select on the bar chart)'); // a detail scatterplot of occurrance vs. position const detail = vl.markPoint({filled: true, opacity:0.6}) .encode( vl.x().fieldQ('Positive Percentage'), //.scale({type: 'log'}), vl.y().fieldQ('Negative Percentage'), //.scale({type: 'sqrt'}), vl.size().fieldQ('Occurrences').scale({range: [10, 300]}) .legend({orient: 'bottom', titleOrient: 'Top'}), vl.opacity().value(0.6), vl.tooltip(['Occurrences','Negative Percentage','Neutral Percentage','Positive Percentage','Emoji']), opacity) // modulate point opacity based on the brush selection .width(300) .height(300)// set chart width,height to 300 pixels .title('The Emotion Every Emoji Conveys') ; // vertically concatenate (vconcat) charts return vl.data(data).vconcat(overview, detail).render(); }3d:T46b,md`

## General Analysis 1. The most commonly used categories are Emoticons, Dingbats, Miscellaneous Symbols, Miscellaneous Symbols and Pictographs.
Although I have cleaned the data, the categories of emojis seem still a little unclear and overlapped with each other, which may to some extent reflect that not a specific category or several of them are very heavily used. People use them not by picking specific categories but by content and icons themselves. 2. Among the top 4 popular categories, their emojis' main contents vary from onne to another.
Using the brush to filter emojis by top4 popular categories in the data vis above, we can see the graph containing data vis as following 4 graphs. There are mainly 2 findings from them.

People use the emojis from Miscellaneous Symbols and Pictographs most.
The senntiment distribution of the emojis in Dingbats, Miscellaneous Symbols, Miscellaneous Symbols and Pictographs most are very similar. Emojis from Emoticons could be more often used to express both positive and negative content .

` 3e:[] 38:{"id":"576c2578ae919fd3","slug":"exploratory-emoji-sentiment-analysis-of-tweets-in-2015","title":"Exploratory Emoji Sentiment Analysis of Tweets in 2015","thumbnail":"a1511a16ef0ebd8e150a1a49c2126da71001f30a1761257bc146f35431fb83b2","access_level":"public","create_time":"$D2020-02-07T22:23:15.020Z","update_time":"$D2020-02-19T07:54:10.886Z","view_time":null,"first_public_time":"$D2020-02-08T22:03:35.872Z","has_unpublished_edits":null,"fork":{"id":"4734cbbb9bcb7391","slug":"anbiotic-data-visualization-design","title":"Anbiotic Data Visualization Design","owner":{"id":"7c9d5fe13c8b86f0","login":"miaxmzhu","name":"Xumin Zhu"}},"stars":0,"starred":false,"owner":{"type":"team","id":"7c9d5fe13c8b86f0","login":"miaxmzhu","name":"Xumin Zhu","avatar_url":"https://avatars-next.observableusercontent.com/39860168d529a9e61725c082710a869217e9a0286913d71cbcb39a1bda1ec332","tier":"starter_2024","bio":"","followed_by_you":false,"follows_you":false},"version":811,"publish_version":811,"role":null,"authored":false,"authors":[{"type":"individual","id":"f8a5c8cdcbd42edf","login":"miaxmzhu","name":"Xumin Zhu","avatar_url":"https://avatars-next.observableusercontent.com/39860168d529a9e61725c082710a869217e9a0286913d71cbcb39a1bda1ec332","tier":null,"bio":"","followed_by_you":false,"follows_you":false}],"members":[{"type":"individual","id":"f8a5c8cdcbd42edf","login":"miaxmzhu","name":"Xumin Zhu","avatar_url":"https://avatars-next.observableusercontent.com/39860168d529a9e61725c082710a869217e9a0286913d71cbcb39a1bda1ec332","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"Exploratory Emoji Sentiment Analysis of Tweets in 2015","cells":[{"id":0,"value":"md`# Exploratory Emoji Sentiment Analysis of Tweets in 2015 \n#IMT562 Assignment 2`\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":21,"value":"md`\nIn this exploratory data analysis, I use an open dataset from Kaggle that has a lexicon of 751 emoji characters with automatically assigned sentiment, computed from 70,000 tweets in 2015. My analysis goal is to understand the quesitons including:
\n1. What emojis and categories of emojis were most popular in 2015?
\n\n2. Among the most popular 30 emojis, what emotions did they convey? Does people always use an emoji to express their positive, neutral or negative moods?\n\n\nOriginal Data from https://www.kaggle.com/thomasseleck/emoji-sentiment-data.`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":50,"value":"md`\n

\n## Import\n\nTo start, I shall import the necessary \\`vega-lite-api\\` and D3 libraries:\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":144,"value":"import {vl} from '@vega/vega-lite-api'","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":220,"value":"import {printTable} from '@uwdata/data-utilities'","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":395,"value":"d3 = require(\"d3@5\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":259,"value":"md`\n

\n## Data\nAnd then, I import the 2 original and cleaned emoji sentiment datasets:\n` \n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":378,"value":"data = d3.csv(\"https://raw.githubusercontent.com/mia-zhu/imt562/master/Emoji_Sentiment_Data_v1.0%20(1)%20(1).csv\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":500,"value":"top30dataformatted = d3.csv(\"https://raw.githubusercontent.com/mia-zhu/imt562/master/top30_Emoji_Sentiment_emotion%20-%20Emoji_Sentiment_data_v2(1).csv\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":733,"value":"top4categories = d3.csv(\"https://raw.githubusercontent.com/mia-zhu/imt562/master/filtered_top4_categories%20-%20Emoji_Sentiment_Data_v1.0%20(1)%20(1)%20(1).csv\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":413,"value":"md`Let's take a peek at the first ten rows of the original dataset:`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":416,"value":"printTable(data.slice(0, 10))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":507,"value":"md`Then let's see the formatted dataset only including top 30 most frequiently used emojis :`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":510,"value":"printTable(top30dataformatted.slice(0, 10))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":279,"value":"md`\n

\n## Exploratory Analysis\nWith the 2 datasets cleaned and wrangled, I start my exploratory analysis. Let's do a recap of my 2 analysis problems--\n1. What emojis and categories of emojis were most popular in 2015?
\n\n2. Among the most popular 30 emojis, what emotions did they convey? Does people always use an emoji to express their positive, neutral or negative moods?\n\nBased on the proposed problems, my method is to analyze from the overall aspect to the focus on top 30 popular emojis.\n\n` \n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":188,"value":"heatmapDiv2 = html`

`;","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":430,"value":"$3c","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":654,"value":"$3d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":685,"value":"vl.markPoint({filled: true})\n .data(top4categories)\n .encode(\n vl.x().fieldQ('Positive Percentage'),\n vl.y().fieldQ('Negative Percentage'),\n vl.size().fieldQ('Occurrences').scale({range: [10, 300]})\n .legend({orient: 'bottom', titleOrient: 'left'}),\n //vl.opacity().value(0.5),\n //vl.tooltip(['Occurrences','Negative Percentage','Neutral Percentage','Positive Percentage','Emoji']),\n // vl.order().fieldQ('Occurrences').sort('descending'),\n vl.column().fieldN('Unicode block').title(null)\n )\n .width(200)\n .height(200)\n //.title('Emoji Contents in Top4 Popular Categories',{ titleOrient: 'middle'})\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":761,"value":"md`\n

\n3.In what position of tweets do people usually use emojis to express different emotions? \n
\nFrom the following graphs, we can see that people usually use emojis in the 2/3 part of their tweets. Take a closer look at the most dense area of each graph. It's interesting that they usually express strong positive emotions, medium neutral attitudes and slight negative moods by using emojis.
\n\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":598,"value":"vl.markPoint()//Circle({size: 15, opacity: 0.5})\n .encode(\n vl.x().fieldQ(vl.repeat('row')),\n vl.y().fieldQ(vl.repeat('column')),\n )\n .width(200)\n .height(200)\n .repeat({\n row: ['Position'],\n column: ['Negative Percentage', 'Positive Percentage', 'Neutral Percentage']\n })\n .data(data)\n \n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":770,"value":"md`\n

\n## Detailed Analysis of Top 30 Emojis\nUnderstanding the general use of emojis, let's take a further step to analyze the top30 popular emojis.\n`\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":464,"value":"vl.markBar({stroke:null,fill:\"pink\"})\n .width(840) \n .data(top30dataformatted)\n.encode(\nvl.x().fieldN(\"Emoji\").sort({field:\"\",order:\"descending\"}).axis({title: 'Top 30 Hottest Emojis', labelAngle: 0}), // no title, no label angle,\n vl.y().fieldQ(\"Percentage\"),\n vl.color().fieldN(\"Emotion\"), //.scale({scheme: 'category20c'}),\n vl.tooltip([\"Percentage\",\"Emotion\",\"Occurrences\",\"Emoji\"]),\n vl.order().fieldQ(\"Emotion\")\n)\n.render()\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":314,"value":"md`We can clearly see that aong the top 30 hottest emojis, most of them are used to express positive emotions instead of negative ones. Those negative 6 emojis people love to use are 😂😭😫😒😔. Some other detailed findings include:
\n

Heart related emojis are very popular among Twitter users and there are 5 heart emojis widely used.
Cry face emojis are not always used to mainly express negative emotions. For example, 😂 conveys positive emotions more often while 😭 speaks out negative moods in more cases.
Personally, I readthe chart and understand that people want to express their positive moods, such as love/ support/happy/smile,in a more straight way; however, to express their negative moods, such as cry/sad/sorry/pitiful/ frustrated, they usually use more subtle ones including 😫😒😔.

`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":789,"value":"md`\n

\n3.Understand The Way People Use Top30 Popular Emojis in Their Tweets \n
\nFrom the followingraphs, we can further understand how people use these top30 emojis.
\n

Most of them are commonly used in the 4/5 position of those tweets.
Most of them convey strong positive or negative moods instead of neutral attitudes.

\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":799,"value":"vl.markPoint({filled: true})\n .data(top30dataformatted)\n .encode(\n vl.x().fieldQ('Position'),\n vl.y().fieldQ('Percentage'),\n vl.size().fieldQ('Occurrences').scale({range: [10, 300]})\n .legend({orient: 'bottom', titleOrient: 'left'}),\n //vl.opacity().value(0.5),\n //vl.tooltip(['Occurrences','Negative Percentage','Neutral Percentage','Positive Percentage','Emoji']),\n // vl.order().fieldQ('Occurrences').sort('descending'),\n vl.column().fieldN('Emotion').title(null),\n vl.tooltip(['Occurrences','Percentage','Emotion','Emoji']),\n )\n .width(200)\n .height(200)\n .title('Understand The Usage of Top30 Popular Emojis ')\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":328,"value":"md`\n

\n## More in-depth data-vis are coming soon!\n

`\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W3e","stdlib":"1"}} 33:null