Introduction

Topic of interest: Tesla Motors, electric car company
Hashtags: '#Tesla', '#TeslaMotors', '#ElonMusk', '#Elon', '#Model3', '#ModelX', '#ModelY', '#TeslaRoadster'

01/0402/0403/0404/0405/0408/0409/0410/0411/0412/0413/0414/0415/0416/0417/0418/0419/0420/0421/0422/0424/0425/0402/0503/0504/0505/0506/0507/0508/0509/0511/0512/0513/0514/0515/0517/0518/0519/0520/0521/0522/0524/0525/0526/0527/0528/0529/0530/050200040006000

Distribution of 154.717 collected tweets over time



In this project we focus on three levels of analysis on twitter collected data:
Emerging Topic Detection, Sentiment Analysis and Geo-location Analysis.

Emerging Topic Detection

In this part of the project, we attempt to extract information about what twitter users talk about in their tweets and if topics of discussion can indicate real-life events.


Word-cloud of 200 most common keywords/hashtags

01/0402/0403/0404/0405/0406/0407/0408/0409/0410/0411/0412/0413/0414/0415/0416/0417/0418/0419/0420/0421/0422/0423/0424/0425/0426/0427/0428/0429/0430/0401/0502/0503/0504/0505/0506/0507/0508/0509/0510/0511/0512/0513/0514/0515/0516/0517/0518/0519/0520/0521/0522/0523/0524/0525/0526/0527/0528/0529/0530/05
teslamodel3elonmusktronbttusdtspacexevteslamodel3trx

Distribution of hashtags over time




List of topics:


01/04/19 - 02/04/19: ['model3 car tsla', 'tesla elonmusk harambe', 'amp']
Elon Musk shares link to his rap song as a tribute to the death of gorilla Harambe: "Emo G Records" 19.495 Retweets

03/04/19 - 05/04/19: ['car tsla', 'goes money work reset knows who help black screen back return']
Tweet of singer @SherylCrow to @Tesla: "Help! Who knows what to do when your @Tesla screen goes black and the reset doesn’t work? Return it and get your money back?? #Tesla #stuckinaparkinglot". 530 Retweets

07/04/19 - 08/04/19: ['10 free cost', 'thank entire sold time lead march cars for team 37 model3 electric market model outsold huge car runner amp all congrats 1st in up breaking history fully']
Tweet of @ceo_plus_ch congratulating Elon Musk: "BREAKING: In 🇨🇭, the #Tesla #Model3 outsold ALL other cars in March. For the 1st time in history, a fully electric car (#BEV) is the most sold model of the entire car market - with a huge lead of 37% over the runner-up. Congrats & thank you, team @Tesla and @elonmusk!". 2.463 Retweets

09/04/19: ['car china', 'this 2019', 'best increase months everyone software thanks getting wow new time the update performance']
Reply by Elon Musk on increasing power and speed performance of Model3 car due to software update: "Firmware update coming later this month will increase power by ~5% & top speed by 10 km/h or 7mph". 215 Retweets

10/04/19: ['best this everyone update thanks vehicles new wow']
Tweet by @jwolfgram on thanking Elon Musk for the software update on model3 car: "WOW!! The @Tesla software update performance increase is like getting a NEW CAR... Again. This is the 3rd time I’ve woken up to a “new” car since I bought #Model3 six months ago. Thanks @elonmusk and everyone at Tesla for making the best vehicles on earth! 💕#AMAZED". 622 Retweets

11/04/19: ['park spacex', 'believe waiting whoever pet parked day refuses sit 100 cal']
Funny tweet by @idiocyafoot: "Cal 100% REFUSES to believe that a car can park itself...and will sit there all day waiting for whoever parked it to get out and pet her. #tesla @elonmusk #goldenretriever #dogs". 1065 Retweets

12/04/19 - 13/04/19: ['money happiness crying buy better lot whole', 'miles 27 over thanks love new thxelon year 000']
Tweet by @cobrarsnake: "Money doesn’t buy happiness... but it is a whole lot better crying in my #Tesla". 470 Retweets
Tweet by @McHoffa: "Over a year and 27,000 miles and I love this car more than when it was new. Thanks @elonmusk @Tesla #Tesla #Model3 #ThxElon". 385 Retweets

16/04/19 - 17/04/19: ['repo impact']
Probably referring to Susan Repo leaving Tesla.

19/04/19 - 20/04/19: ['amp tron it usdt cash btt 20m coming news success celebrate airdrop planning to free bad good i']
Tweet by @justinsuntron on giveaway: "To celebrate #BTT & #USDT-#TRON success, I am planning a $20m free cash airdrop. Good news-it's coming, bad news-I may decide to give away more! First, I will randomly pick 1 winner for a #Tesla up until 3/27! To apply, follow me and RT this tweet! Simple! #Blockchain". 50.723 Retweets

21/04/19 - 22/04/19: ['tron cash coming 20m news airdrop planning free usdt success amp it btt celebrate', 'this bad shanghai happened evs china anything chi positive i today post to negative good']
Deleted tweet by @ShanghaiJayin: "Good or bad, negative or positive I will post anything about Tesla or EVs in China. This happened today in Shanghai, China 🇨🇳 1st generation Tesla Model S caught Fire 🔥 underground car park.#Tesla #TeslaChina #ModelS #Fire #China #Shanghai #特斯拉 #中国 $TSLA pic.twitter.com/HOwMcvulV1".

25/04/19: ['q1', 'tsla insurance']
Tesla on introducing its own insurance product after q1: "Tesla disrupting the car insurance business. That’s what you can do when you have all the data (and a safe car). Nice side effect: generating some float. $TSLA".

03/05/19 - 04/05/19: ['first full computer four vehicles ahead learn competition latest research years self according launch production driving']
Tweet by @ARKInvest: "#Tesla’s full self-driving computer is the first such computer to launch in production vehicles and, according to our research, is four years ahead of the competition.". 852 Retweets

05/05/19: ['police cruiser acceleratepolicing model3']
Tesla Model 3 police car makes an appearance at law enforcement tech conference: "Well done @axon_us @ashleysanichar @AxonRick for hosting a world-class #AcceleratePolicing conference in Phoenix Arizona USA. Technology is indeed the future and let’s hope that cops the world over embrace it. @CSIWorld @SAPoliceService @GP_CommSafety @Yoliswamakhasi #CrimeWatch".

06/05/19 - 07/05/19: ['monumental years made ed achievements video 17 ago here spacex founded celebrating sta']
Elon Musk sharing video link of cargo craft from @Space_Station: "Dragon approaching Space Station". 2.724 Retweets

09/05/19: ['impressed lane coming time autonomy tsla change nav seriously first disbelieves needs wow try anyone auto amazing using ap']
Tweet by @kimpaquette: "Wow. @elonmusk First time using Nav on AP with auto lane change and it is AMAZING. Anyone who disbelieves autonomy is coming needs to try it out. I’m seriously impressed. #Tesla @tesla #myfuturerobotaxi $TSLA". 620 Retweets

11/05/19 - 12/05/19: ['please finger request work saturation undos multiple picker add painting color adjustment', 'just world software check come']
Tweet by @gorosart: "Tesla finger painting. A request to @elonmusk: please add a color picker model with saturation adjustment and multiple undos...how else am I supposed to do serious freelance work with this? 😂 #Tesla". 529 Retweets

13/05/19: ['world software', '000 28 winner usd trx']
Tweet by @justinsuntron announcing giveaway winner: "I'm also glad to announce the 2nd winner @uzgaroth of #Tesla giveaway. 28,000 $USD value of #TRX had been transferred to his wallet, all the documents were compliant. #TRON keeps up to the promises and keeps delivering the results. Stay with us for more exciting news on the way.". 295 Retweets

24/05/19 - 27/05/19: ['this things coolest dominance growth one ev tsla show']
Tweet by @GerberKawasaki: "This could be one of the coolest things to show Tesla’s EV dominance and the #Model3 growth. $tsla". 834 Retweets

28/05/19: ['video']
Tweet by @TeslaNY: "The Data Proving #Tesla's Success – Video 🚀📊 https://youtu.be/XkKyeSiWMe8 $TSLA #EV #Model3 @elonmusk". 207 Retweets

Sentiment Analysis

In this part of the project, we are called to identify and categorize opinions expressed in tweets, and better understand the users' emotions over Tesla. We followed two approaches, one to classify tweets in the scale of positive, neutral and negative, and a second one to match tweets on specific emotions:
anger, joy, disgust, fear, sadness, surprise.


Overall Sentiment

01/0402/0403/0404/0405/0406/0407/0408/0409/0410/0411/0412/0413/0414/0415/0416/0417/0418/0419/0420/0421/0422/0423/0424/0425/0426/0427/0428/0429/0430/0401/0502/0503/0504/0505/0506/0507/0508/0509/0510/0511/0512/0513/0514/0515/0516/0517/0518/0519/0520/0521/0522/0523/0524/0525/0526/0527/0528/0529/0530/050200040006000
negativeneutralpositive

Sentiment of tweets over time

01/0402/0403/0404/0405/0406/0407/0408/0409/0410/0411/0412/0413/0414/0415/0416/0417/0418/0419/0420/0421/0422/0423/0424/0425/0426/0427/0428/0429/0430/0401/0502/0503/0504/0505/0506/0507/0508/0509/0510/0511/0512/0513/0514/0515/0516/0517/0518/0519/0520/0521/0522/0523/0524/0525/0526/0527/0528/0529/0530/05−0.2−0.100.1

Polarity of tweets over time

01/0402/0403/0404/0405/0406/0407/0408/0409/0410/0411/0412/0413/0414/0415/0416/0417/0418/0419/0420/0421/0422/0423/0424/0425/0426/0427/0428/0429/0430/0401/0502/0503/0504/0505/0506/0507/0508/0509/0510/0511/0512/0513/0514/0515/0516/0517/0518/0519/0520/0521/0522/0523/0524/0525/0526/0527/0528/0529/0530/05
angerjoydisgustfearsadnesssurprise

Sentiment value over time



Specific Emotions



Word-cloud of most common keywords/hashtags in "anger" tweets



Word-cloud of most common keywords/hashtags in "joy" tweets



Word-cloud of most common keywords/hashtags in "fear" tweets



Word-cloud of most common keywords/hashtags in "sadness" tweets



Word-cloud of most common keywords/hashtags in "surprise" tweets



Word-cloud of most common keywords/hashtags in "disgust" tweets



Geo-location

In this part of the project, our task is to extract information of users' geographic locations based on their tweets.

Location and sentiment of tweets



Word-cloud of the most common tweet locations