Who Will Win the 2016 Presidential Election?
## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## 1 -1 0.6666667 0.0000000 0.6666667
## 2 1 0.5000000 0.5000000 0.0000000
## 3 0 0.3333333 0.1666667 0.1666667
## senti_diffs_per_ref
## 1 -0.6666667
## 2 0.5000000
## 3 0.0000000## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## 1 -1 0.5 0.0 0.5
## 2 1 0.5 0.5 0.0
## senti_diffs_per_ref
## 1 -0.5
## 2 0.5## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## 1 1.00000000 0.09803922 0.09803922 0.00000000
## 2 0.09090909 0.25581395 0.13953488 0.11627907
## 3 0.50000000 0.12500000 0.09375000 0.03125000
## 4 1.00000000 0.07317073 0.07317073 0.00000000
## 5 1.00000000 0.08333333 0.08333333 0.00000000
## 6 0.60000000 0.07352941 0.05882353 0.01470588
## 7 -0.14285714 0.14893617 0.06382979 0.08510638
## 8 -1.00000000 0.06666667 0.00000000 0.06666667
## 9 1.00000000 0.06122449 0.06122449 0.00000000
## 10 0.25000000 0.23529412 0.14705882 0.08823529
## 11 0.00000000 0.10526316 0.05263158 0.05263158
## 12 0.00000000 0.05555556 0.02777778 0.02777778
## 13 0.60000000 0.09259259 0.07407407 0.01851852
## 14 1.00000000 0.03030303 0.03030303 0.00000000
## 15 -0.33333333 0.09375000 0.03125000 0.06250000
## 16 1.00000000 0.03703704 0.03703704 0.00000000
## 17 0.00000000 0.07547170 0.03773585 0.03773585
## 18 0.60000000 0.17857143 0.14285714 0.03571429
## 19 0.33333333 0.11111111 0.07407407 0.03703704
## 20 0.00000000 0.00000000 0.00000000 0.00000000
## 21 1.00000000 0.06250000 0.06250000 0.00000000
## 22 1.00000000 0.10000000 0.10000000 0.00000000
## 23 -0.60000000 0.19230769 0.03846154 0.15384615
## 24 0.60000000 0.15151515 0.12121212 0.03030303
## 25 1.00000000 0.12500000 0.12500000 0.00000000
## 26 0.00000000 0.07142857 0.03571429 0.03571429
## 27 0.00000000 0.21052632 0.10526316 0.10526316
## 28 0.33333333 0.09090909 0.06060606 0.03030303
## 29 0.00000000 0.06451613 0.03225806 0.03225806
## 30 -0.33333333 0.10714286 0.03571429 0.07142857
## senti_diffs_per_ref
## 1 0.09803922
## 2 0.02325581
## 3 0.06250000
## 4 0.07317073
## 5 0.08333333
## 6 0.04411765
## 7 -0.02127660
## 8 -0.06666667
## 9 0.06122449
## 10 0.05882353
## 11 0.00000000
## 12 0.00000000
## 13 0.05555556
## 14 0.03030303
## 15 -0.03125000
## 16 0.03703704
## 17 0.00000000
## 18 0.10714286
## 19 0.03703704
## 20 0.00000000
## 21 0.06250000
## 22 0.10000000
## 23 -0.11538462
## 24 0.09090909
## 25 0.12500000
## 26 0.00000000
## 27 0.00000000
## 28 0.03030303
## 29 0.00000000
## 30 -0.03571429## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.0000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.: 0.0000 1st Qu.:0.06786 1st Qu.:0.03604 1st Qu.:0.00000
## Median : 0.3333 Median :0.09317 Median :0.06186 Median :0.03078
## Mean : 0.3499 Mean :0.10588 Mean :0.06811 Mean :0.03778
## 3rd Qu.: 1.0000 3rd Qu.:0.12500 3rd Qu.:0.09697 3rd Qu.:0.06003
## Max. : 1.0000 Max. :0.25581 Max. :0.14706 Max. :0.15385
## senti_diffs_per_ref
## Min. :-0.11538
## 1st Qu.: 0.00000
## Median : 0.03367
## Mean : 0.03033
## 3rd Qu.: 0.06250
## Max. : 0.12500## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-0.3333 Min. :0.03030 Min. :0.02778 Min. :0.00000
## 1st Qu.: 0.0000 1st Qu.:0.06970 1st Qu.:0.03571 1st Qu.:0.00000
## Median : 0.3333 Median :0.09175 Median :0.06720 Median :0.03030
## Mean : 0.4296 Mean :0.10619 Mean :0.07334 Mean :0.03285
## 3rd Qu.: 1.0000 3rd Qu.:0.11458 3rd Qu.:0.10383 3rd Qu.:0.03721
## Max. : 1.0000 Max. :0.25581 Max. :0.14706 Max. :0.11628
## senti_diffs_per_ref
## Min. :-0.03571
## 1st Qu.: 0.00000
## Median : 0.03367
## Mean : 0.04049
## 3rd Qu.: 0.06421
## Max. : 0.12500## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.0000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.:-0.1071 1st Qu.:0.06838 1st Qu.:0.03739 1st Qu.:0.00000
## Median : 0.2500 Median :0.09688 Median :0.06066 Median :0.03348
## Mean : 0.2588 Mean :0.10553 Mean :0.06212 Mean :0.04341
## 3rd Qu.: 0.9000 3rd Qu.:0.14295 3rd Qu.:0.09115 3rd Qu.:0.06563
## Max. : 1.0000 Max. :0.21053 Max. :0.14286 Max. :0.15385
## senti_diffs_per_ref
## Min. :-0.11538
## 1st Qu.:-0.01596
## Median : 0.01852
## Mean : 0.01872
## 3rd Qu.: 0.06250
## Max. : 0.10714## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :0.00000 Min. :0.06452 Min. :0.03226 Min. :0.03030
## 1st Qu.:0.08333 1st Qu.:0.07111 1st Qu.:0.03935 1st Qu.:0.03079
## Median :0.16667 Median :0.07771 Median :0.04643 Median :0.03128
## Mean :0.16667 Mean :0.07771 Mean :0.04643 Mean :0.03128
## 3rd Qu.:0.25000 3rd Qu.:0.08431 3rd Qu.:0.05352 3rd Qu.:0.03177
## Max. :0.33333 Max. :0.09091 Max. :0.06061 Max. :0.03226
## senti_diffs_per_ref
## Min. :0.000000
## 1st Qu.:0.007576
## Median :0.015152
## Mean :0.015152
## 3rd Qu.:0.022727
## Max. :0.030303## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.0000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.: 0.0000 1st Qu.:0.07186 1st Qu.:0.03792 1st Qu.:0.00000
## Median : 0.4167 Median :0.09589 Median :0.06850 Median :0.03078
## Mean : 0.3653 Mean :0.11065 Mean :0.07222 Mean :0.03843
## 3rd Qu.: 1.0000 3rd Qu.:0.14295 3rd Qu.:0.09951 3rd Qu.:0.06003
## Max. : 1.0000 Max. :0.25581 Max. :0.14706 Max. :0.15385
## senti_diffs_per_ref
## Min. :-0.11538
## 1st Qu.: 0.00000
## Median : 0.04058
## Mean : 0.03378
## 3rd Qu.: 0.07050
## Max. : 0.12500## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-0.3333 Min. :0.03704 Min. :0.03571 Min. :0.00000
## 1st Qu.: 0.0000 1st Qu.:0.05456 1st Qu.:0.03604 1st Qu.:0.01786
## Median : 0.3333 Median :0.07209 Median :0.03638 Median :0.03571
## Mean : 0.3333 Mean :0.07209 Mean :0.03638 Mean :0.03571
## 3rd Qu.: 0.6667 3rd Qu.:0.08962 3rd Qu.:0.03671 3rd Qu.:0.05357
## Max. : 1.0000 Max. :0.10714 Max. :0.03704 Max. :0.07143
## senti_diffs_per_ref
## Min. :-0.0357143
## 1st Qu.:-0.0175265
## Median : 0.0006614
## Mean : 0.0006614
## 3rd Qu.: 0.0188492
## Max. : 0.0370370
## [1] "Using direct authentication"## [1] 1## [1] "Rate limited .... blocking for a minute and retrying up to 119 times ..."## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.00 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.: 0.00 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000
## Median : 0.00 Median :0.06066 Median :0.00000 Median :0.00000
## Mean : 0.01 Mean :0.07348 Mean :0.03633 Mean :0.03715
## 3rd Qu.: 0.00 3rd Qu.:0.12500 3rd Qu.:0.07280 3rd Qu.:0.07280
## Max. : 1.00 Max. :0.33333 Max. :0.16667 Max. :0.21429
## senti_diffs_per_ref
## Min. :-0.1428571
## 1st Qu.: 0.0000000
## Median : 0.0000000
## Mean :-0.0008273
## 3rd Qu.: 0.0000000
## Max. : 0.1538461
By comparing the top two Democratic and Republican candidates, I predict that Doanld Trump will win the election because he has the second highest network density, the most mentions on twitter around Washington, realtively equal positive and neutral remarks composing tweets he is mentioned in, and positive mean polarity and mean sentiment.
## [1] "Using direct authentication"## [1] 1## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.0000 Min. :0.00000 Min. :0.00000 Min. :0.0000
## 1st Qu.: 0.0000 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.0000
## Median : 0.0000 Median :0.07143 Median :0.00000 Median :0.0000
## Mean : 0.1144 Mean :0.07288 Mean :0.04388 Mean :0.0290
## 3rd Qu.: 1.0000 3rd Qu.:0.12500 3rd Qu.:0.07692 3rd Qu.:0.0625
## Max. : 1.0000 Max. :0.46154 Max. :0.40000 Max. :0.2500
## senti_diffs_per_ref
## Min. :-0.25000
## 1st Qu.: 0.00000
## Median : 0.00000
## Mean : 0.01489
## 3rd Qu.: 0.06667
## Max. : 0.40000## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.0000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.: 0.0000 1st Qu.:0.05000 1st Qu.:0.00000 1st Qu.:0.00000
## Median : 0.3333 Median :0.08333 Median :0.06458 Median :0.00000
## Mean : 0.1801 Mean :0.10513 Mean :0.06938 Mean :0.03575
## 3rd Qu.: 1.0000 3rd Qu.:0.20000 3rd Qu.:0.13333 3rd Qu.:0.06667
## Max. : 1.0000 Max. :0.55556 Max. :0.55556 Max. :0.20000
## senti_diffs_per_ref
## Min. :-0.20000
## 1st Qu.: 0.00000
## Median : 0.05556
## Mean : 0.03362
## 3rd Qu.: 0.06667
## Max. : 0.55556## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.00000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.:-0.33333 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000
## Median : 0.00000 Median :0.07692 Median :0.00000 Median :0.00000
## Mean : 0.02102 Mean :0.09039 Mean :0.04466 Mean :0.04573
## 3rd Qu.: 0.60000 3rd Qu.:0.13333 3rd Qu.:0.07692 3rd Qu.:0.07692
## Max. : 1.00000 Max. :0.75000 Max. :0.50000 Max. :0.40000
## senti_diffs_per_ref
## Min. :-0.400000
## 1st Qu.:-0.052632
## Median : 0.000000
## Mean :-0.001066
## 3rd Qu.: 0.066667
## Max. : 0.400000## polarity subjectivity pos_refs_per_ref neg_refs_per_ref
## Min. :-1.00000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.: 0.00000 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000
## Median : 0.00000 Median :0.08333 Median :0.00000 Median :0.00000
## Mean : 0.03962 Mean :0.08621 Mean :0.04855 Mean :0.03766
## 3rd Qu.: 0.50000 3rd Qu.:0.12500 3rd Qu.:0.07692 3rd Qu.:0.07692
## Max. : 1.00000 Max. :0.36364 Max. :0.28571 Max. :0.25000
## senti_diffs_per_ref
## Min. :-0.25000
## 1st Qu.: 0.00000
## Median : 0.00000
## Mean : 0.01089
## 3rd Qu.: 0.06250
## Max. : 0.28571polarityMeans = c(-0.08503, 0.337, 0.03197, 0.2528)
barplot(polarityMeans, main="Comparing Mean Polarities of Top Presidential Candidates", names.arg=c("Hillary Clinton", "Bernie Sanders", "Donald Trump", "Ben Carson"))
Bernie Sanders appears to have the most positive remarks made about him, while most remarks about Hillary Clinton seem to lean towards the negative side of things. Ben Carson was also the subject of mostly positive remarks, while thoughts on Donald Trump seem to almost be neutral.
sentimentMeans = c(-0.009639, 0.02218, 0.007444, 0.03925)
barplot(sentimentMeans, main="Comparing Mean Sentiments Towards Top Presidential Candidates", names.arg=c("Hillary Clinton", "Bernie Sanders", "Donald Trump", "Ben Carson"))
In this comparison, Ben Carson had the most positive remarks in relation to negative ones. On the other hand, negative remarks made about Hillary Clinton seem to outnumber the positive ones.
Words Most Commonly Associated with Hillary Clinton
comparison.cloud(ctdm, colors = brewer.pal(3, "Dark2"), scale = c(3,.5),random.order = FALSE)
Hillary’s wordcloud seems to be equal parts positive, negative, and neutral. Just like in Bernie’s below, negative remarks focus around “raise,” while her positive remarks focus around her “strategy.”
Words Most Commonly Associated with Bernie Sanders
comparison.cloud(stdm, colors = brewer.pal(3, "Dark2"), scale = c(3,.5),random.order = FALSE)
In Sanders’ wordcloud, the amount of negative and neutral words seems to be about equal. Remarks focused on topics surrounding the words “raise,” “untold,” and “1987.”
Words Most Commonly Associated with Donald Trump
comparison.cloud(ttdm, colors = brewer.pal(3, "Dark2"), scale = c(3,.5),random.order = FALSE)
As you can see, “hate” is the most commonly used negative word associated with Trump and his campaign. However, it is used significantly less than other positive and neutral words such as “military” and “like.”
Words Most Commonly Associated with Ben Carson
comparison.cloud(btdm, colors = brewer.pal(3, "Dark2"), scale = c(3,.5),random.order = FALSE)
Carson’s wordcloud had the most reused words out of all his competitors. Positive remarks surround topics like “refugees” and “terrorists.” Neutral and negative remarks take a broader view and involve frequently used words including “wellness,” “poll,” and “meek.”
Hillary Clinton’s Support Around Washington, D.C.
qmplot(longitude, latitude, data = cgeoDF_noNA, colour = I('red'),source="google")
Bernie Sanders’ Support Around Washington, D.C.
qmplot(longitude, latitude, data = sgeoDF_noNA, colour = I('red'),source="google")
Donald Trump’s Support Around Washington, D.C.
qmplot(longitude, latitude, data = tgeoDF_noNA, colour = I('red'),source="google")
Ben Carson’s Support Around Washington, D.C.
qmplot(longitude, latitude, data = bgeoDF_noNA, colour = I('red'),source="google")
When looking over all the candidates’geographical support maps, we see that Trump was tweeted about the most around the D.C. area and Sanders was tweeted about the least. Mentions of Clinton seem to focus more on the inner cities of D.C. and Baltimore as opposed to her competitors mentions, whose also include the outlying suburbs.
## [1] "Using direct authentication"## [1] 1The Twitter Network of Hillary Clinton
plot(MyGraph,
layout=layout.fruchterman.reingold,
main='Twitter Network',
vertex.color = 'green',
vertex.label.dist=0.25,
vertex.frame.color='blue',
vertex.label.color='black',
vertex.label.font=0.1,
vertex.label.cex=0.5,
vertex.size = 0.5,
vertex.shape = "circle",
edge.arrow.size = 0,
edge.lty = 6,
edge.curved=0.5
)
The Twitter Network of Bernie Sanders
plot(sMyGraph,
layout=layout.fruchterman.reingold,
main='Twitter Network',
vertex.color = 'green',
vertex.label.dist=0.25,
vertex.frame.color='blue',
vertex.label.color='black',
vertex.label.font=0.1,
vertex.label.cex=0.5,
vertex.size = 0.5,
vertex.shape = "circle",
edge.arrow.size = 0,
edge.lty = 6,
edge.curved=0.5
)
The Twitter Network of Donald Trump
plot(tMyGraph,
layout=layout.fruchterman.reingold,
main='Twitter Network',
vertex.color = 'green',
vertex.label.dist=0.25,
vertex.frame.color='blue',
vertex.label.color='black',
vertex.label.font=0.1,
vertex.label.cex=0.5,
vertex.size = 0.5,
vertex.shape = "circle",
edge.arrow.size = 0,
edge.lty = 6,
edge.curved=0.5
)
The Twitter Network of Ben Carson
plot(bMyGraph,
layout=layout.fruchterman.reingold,
main='Twitter Network',
vertex.color = 'green',
vertex.label.dist=0.25,
vertex.frame.color='blue',
vertex.label.color='black',
vertex.label.font=0.1,
vertex.label.cex=0.5,
vertex.size = 0.5,
vertex.shape = "circle",
edge.arrow.size = 0,
edge.lty = 6,
edge.curved=0.5
)
## [1] 0## [1] 0.009679371## [1] 0## [1] 0.01585624## [1] 0## [1] 0.008928571## [1] 0## [1] 0.008751903networkDensities = c(0.01295097, 0.010181, 0.01245791, 0.008249497)
barplot(networkDensities, main="Comparing Network Densities of Top Presidential Candidates", names.arg=c("Hillary Clinton", "Bernie Sanders", "Donald Trump", "Ben Carson"))
When comparing candidates’ twitter networks and the network densities, Hillary Clinton has the most dense network of mentions about her and her campaign, with Donald Trump close behind. *Analysis may be a little off because the tweets change daily and therefore affect what words appear in the word clouds.