| none | light | medium | heavy | |
|---|---|---|---|---|
| SM | 4 | 2 | 3 | 2 |
| JM | 4 | 3 | 7 | 4 |
| SE | 25 | 10 | 12 | 4 |
| JE | 18 | 24 | 33 | 13 |
| SC | 10 | 6 | 7 | 2 |
Correspondence analysis
Methods of Corpus Linguistics (class 7)
Mariana Montes
Outline
- Introduction
- Terminology
- Interpretation
- Linguistic example
Correspondence Analysis
Dimension reduction technique for “count” data.
Visualization: biplot, showing relationships:
between rows
between columns
between rows and columns (kind of).
Introduction
Introduction
Linguistic example
Introduction
Classic (non-linguistic) example1
Rows are types of employees (Senior/Junior manager, Senior/Junior Employee, Secretary).
Columns are types of smokers.
Values in the cells are counts.
Introduction
Smoke plot
Introduction
Smoke CA
Principal inertias (eigenvalues):
1 2 3
Value 0.074759 0.010017 0.000414
Percentage 87.76% 11.76% 0.49%
Rows:
SM JM SE JE SC
Mass 0.05699 0.0933 0.2642 0.4560 0.12953
ChiDist 0.21656 0.3569 0.3808 0.2400 0.21617
Inertia 0.00267 0.0119 0.0383 0.0263 0.00605
Dim. 1 -0.24054 0.9471 -1.3920 0.8520 -0.73546
Dim. 2 -1.93571 -2.4310 -0.1065 0.5769 0.78843
Columns:
none light medium heavy
Mass 0.3161 0.23316 0.3212 0.1295
ChiDist 0.3945 0.17400 0.1981 0.3551
Inertia 0.0492 0.00706 0.0126 0.0163
Dim. 1 -1.4385 0.36375 0.7180 1.0744
Dim. 2 -0.3047 1.40943 0.0735 -1.9760
Principal inertias (eigenvalues):
dim value % cum% scree plot
1 0.074759 87.8 87.8 **********************
2 0.010017 11.8 99.5 ***
3 0.000414 0.5 100.0
-------- -----
Total: 0.085190 100.0
Rows:
name mass qlt inr k=1 cor ctr k=2 cor ctr
1 | SM | 57 893 31 | -66 92 3 | -194 800 214 |
2 | JM | 93 991 139 | 259 526 84 | -243 465 551 |
3 | SE | 264 1000 450 | -381 999 512 | -11 1 3 |
4 | JE | 456 1000 308 | 233 942 331 | 58 58 152 |
5 | SC | 130 999 71 | -201 865 70 | 79 133 81 |
Columns:
name mass qlt inr k=1 cor ctr k=2 cor ctr
1 | none | 316 1000 577 | -393 994 654 | -30 6 29 |
2 | lght | 233 984 83 | 99 327 31 | 141 657 463 |
3 | medm | 321 983 148 | 196 982 166 | 7 1 2 |
4 | hevy | 130 995 192 | 294 684 150 | -198 310 506 |Introduction
Terminology
Original matrix \(N\)
\(n \times m\) matrix with frequency counts \(n_{ij}\)
- \(n\) = 5
- \(m\) = 4
- \(n_{\mathrm{SE, light}}\) = 10
Terminology
Correspondence matrix \(P\)
\(n \times m\) matrix with overall proportions \(p_{ij}\)
Mass
\(p_{\mathrm{SE,light}}\) = 0.052
\(p_{\mathrm{SE,.}}\) = 0.264
\(p_{\mathrm{.,light}}\) = 0.233
Terminology
Row profiles
\(n \times m\) matrix with overall row proportions \(r_{ij}\)
\(r_{\mathrm{SE,light}}\) = 0.1961
columns are dimensions in the row points cloud
In bold: the row centroid (vector of column masses!)
Terminology
What do you mean, row point cloud?
Code
row_cloud <- addmargins(smoke_mtx, 1) %>% prop.table(1) %>%
as_tibble(rownames = "Employee") %>%
mutate(Employee = if_else(Employee == "Sum", "Centroid", Employee))
row_cloud %>% ggplot(aes(x = none, y = light, label = Employee)) +
geom_text(size = 10) +
theme_minimal(base_size = 20) +
annotate("point", x = row_cloud$none[[nrow(row_cloud)]], y = row_cloud$light[[nrow(row_cloud)]], shape = 1, size = 25)
BUT more than two dimensions
We then compute \(\chi^2\) distances instead of euclidean distances
How far are points from the centroid?
Terminology
Column profiles
\(n \times m\) matrix with overall row proportions \(c_{ij}\)
\(c_{\mathrm{SE,light}}\) = 0.2221
rows are dimensions in the column points cloud
In bold: the column centroid (vector of row masses!)
Terminology
Also a column point cloud?
Code
row_cloud %>% ggplot(aes(x = none, y = light, label = Employee)) +
geom_text(size = 8) +
theme_minimal(base_size = 20) +
annotate("point", x = row_cloud$none[[nrow(row_cloud)]], y = row_cloud$light[[nrow(row_cloud)]], shape = 1, size = 25) +
labs(title = "Two dimensions of the row cloud.")
col_cloud <- addmargins(smoke_mtx, 2) %>% prop.table(2) %>%
t() %>%
as_tibble(rownames = "Smoker") %>%
mutate(Smoker = if_else(Smoker == "Sum", "Centroid", Smoker))
col_cloud %>% ggplot(aes(x = SM, y = SC, label = Smoker)) +
geom_text(size = 8) +
theme_minimal(base_size = 20) +
annotate("point", x = col_cloud$SM[[nrow(col_cloud)]], y = col_cloud$SC[[nrow(col_cloud)]], shape = 1, size = 25) +
labs(title = "Two dimensions of the column cloud.")
Terminology
Interpretation
Smoke CA - interpretation
Principal inertias (eigenvalues):
1 2 3
Value 0.074759 0.010017 0.000414
Percentage 87.76% 11.76% 0.49%
Rows:
SM JM SE JE SC
Mass 0.05699 0.0933 0.2642 0.4560 0.12953
ChiDist 0.21656 0.3569 0.3808 0.2400 0.21617
Inertia 0.00267 0.0119 0.0383 0.0263 0.00605
Dim. 1 -0.24054 0.9471 -1.3920 0.8520 -0.73546
Dim. 2 -1.93571 -2.4310 -0.1065 0.5769 0.78843
Columns:
none light medium heavy
Mass 0.3161 0.23316 0.3212 0.1295
ChiDist 0.3945 0.17400 0.1981 0.3551
Inertia 0.0492 0.00706 0.0126 0.0163
Dim. 1 -1.4385 0.36375 0.7180 1.0744
Dim. 2 -0.3047 1.40943 0.0735 -1.9760Interpretation
Let’s look at rowmass
Code
| none | light | medium | heavy | |
|---|---|---|---|---|
| SM | 0.364 | 0.182 | 0.273 | 0.182 |
| JM | 0.222 | 0.167 | 0.389 | 0.222 |
| SE | 0.490 | 0.196 | 0.235 | 0.078 |
| JE | 0.205 | 0.273 | 0.375 | 0.148 |
| SC | 0.400 | 0.240 | 0.280 | 0.080 |
| Sum | 0.316 | 0.233 | 0.321 | 0.130 |
| none | light | medium | heavy | Sum | |
|---|---|---|---|---|---|
| SM | 0.066 | 0.044 | 0.048 | 0.08 | 0.057 |
| JM | 0.066 | 0.067 | 0.113 | 0.16 | 0.093 |
| SE | 0.410 | 0.222 | 0.194 | 0.16 | 0.264 |
| JE | 0.295 | 0.533 | 0.532 | 0.52 | 0.456 |
| SC | 0.164 | 0.133 | 0.113 | 0.08 | 0.130 |
[1] 0.316 0.233 0.321 0.130[1] 0.0570 0.0933 0.2642 0.4560 0.1295Interpretation
Let’s look at rowdist
none light medium heavy
0.364 0.182 0.273 0.182 [1] 0.316 0.233 0.321 0.130 none light medium heavy
0.0476 -0.0513 -0.0485 0.0523 Interpretation
Let’s look at rowdist
none light medium heavy
0.364 0.182 0.273 0.182 [1] 0.316 0.233 0.321 0.130 none light medium heavy
0.0476 -0.0513 -0.0485 0.0523 none light medium heavy
0.00226 0.00264 0.00235 0.00273 Interpretation
Let’s look at rowdist
none light medium heavy
0.364 0.182 0.273 0.182 [1] 0.316 0.233 0.321 0.130 none light medium heavy
0.0476 -0.0513 -0.0485 0.0523 none light medium heavy
0.00226 0.00264 0.00235 0.00273 [1] 0.0999Interpretation
Let’s look at rowdist
none light medium heavy
0.364 0.182 0.273 0.182 [1] 0.316 0.233 0.321 0.130 none light medium heavy
0.0476 -0.0513 -0.0485 0.0523 none light medium heavy
0.00226 0.00264 0.00235 0.00273 none light medium heavy
0.00716 0.01131 0.00733 0.02110 # Chi-squared distance
sqrt(sum((prop.table(smoke_mtx,1)["SM",]-smoke_ca$colmass)^2/smoke_ca$colmass))[1] 0.217Interpretation
Let’s look at rowdist
row_dists <- map_dbl(smoke_ca$rownames, function(row) {
sqrt(sum((prop.table(smoke_mtx,1)[row,]-smoke_ca$colmass)^2/smoke_ca$colmass))
})
row_dists[1] 0.217 0.357 0.381 0.240 0.216[1] 0.217 0.357 0.381 0.240 0.216Interpretation
Let’s look at rowinertia
[1] 0.217 0.357 0.381 0.240 0.216[1] 0.0469 0.1274 0.1450 0.0576 0.0467[1] 0.00267 0.01188 0.03831 0.02627 0.00605[1] 0.00267 0.01188 0.03831 0.02627 0.00605Interpretation
Reading inertias
Principal inertias (eigenvalues):
1 2 3
Value 0.074759 0.010017 0.000414
Percentage 87.76% 11.76% 0.49%
Rows:
SM JM SE JE SC
Mass 0.05699 0.0933 0.2642 0.4560 0.12953
ChiDist 0.21656 0.3569 0.3808 0.2400 0.21617
Inertia 0.00267 0.0119 0.0383 0.0263 0.00605
Dim. 1 -0.24054 0.9471 -1.3920 0.8520 -0.73546
Dim. 2 -1.93571 -2.4310 -0.1065 0.5769 0.78843
Columns:
none light medium heavy
Mass 0.3161 0.23316 0.3212 0.1295
ChiDist 0.3945 0.17400 0.1981 0.3551
Inertia 0.0492 0.00706 0.0126 0.0163
Dim. 1 -1.4385 0.36375 0.7180 1.0744
Dim. 2 -0.3047 1.40943 0.0735 -1.9760Interpretation
Interpretation of summary()
Principal inertias (eigenvalues):
dim value % cum% scree plot
1 0.074759 87.8 87.8 **********************
2 0.010017 11.8 99.5 ***
3 0.000414 0.5 100.0
-------- -----
Total: 0.085190 100.0
Rows:
name mass qlt inr k=1 cor ctr k=2 cor ctr
1 | SM | 57 893 31 | -66 92 3 | -194 800 214 |
2 | JM | 93 991 139 | 259 526 84 | -243 465 551 |
3 | SE | 264 1000 450 | -381 999 512 | -11 1 3 |
4 | JE | 456 1000 308 | 233 942 331 | 58 58 152 |
5 | SC | 130 999 71 | -201 865 70 | 79 133 81 |
Columns:
name mass qlt inr k=1 cor ctr k=2 cor ctr
1 | none | 316 1000 577 | -393 994 654 | -30 6 29 |
2 | lght | 233 984 83 | 99 327 31 | 141 657 463 |
3 | medm | 321 983 148 | 196 982 166 | 7 1 2 |
4 | hevy | 130 995 192 | 294 684 150 | -198 310 506 |round(smoke_ca$rowmass*1000)
round(smoke_ca$rowinertia/sum(smoke_ca$rowinertia)*1000)
round(smoke_ca$colinertia/sum(smoke_ca$colinertia)*1000)[1] 57 93 264 456 130[1] 31 139 450 308 71[1] 577 83 148 192:::
::::
Interpretation
Zooming in on rows
rows <- summary(smoke_ca)$rows
colnames(rows) <- c("row", "mass", "quality", "inertia",
paste0(rep(c("k", "cor", "ctr"), 2), rep(c(1, 2), each = 3)))
rows <- as_tibble(rows)
# left table
rows %>% print_matrix()
# right table
rows %>% select(row, quality, cor1, cor2) %>%
mutate(cors = cor1+cor2, qdiff = quality-cors) %>%
print_matrix()| row | mass | quality | inertia | k1 | cor1 | ctr1 | k2 | cor2 | ctr2 |
|---|---|---|---|---|---|---|---|---|---|
| SM | 57 | 893 | 31 | -66 | 92 | 3 | -194 | 800 | 214 |
| JM | 93 | 991 | 139 | 259 | 526 | 84 | -243 | 465 | 551 |
| SE | 264 | 1000 | 450 | -381 | 999 | 512 | -11 | 1 | 3 |
| JE | 456 | 1000 | 308 | 233 | 942 | 331 | 58 | 58 | 152 |
| SC | 130 | 999 | 71 | -201 | 865 | 70 | 79 | 133 | 81 |
| row | quality | cor1 | cor2 | cors | qdiff |
|---|---|---|---|---|---|
| SM | 893 | 92 | 800 | 892 | 1 |
| JM | 991 | 526 | 465 | 991 | 0 |
| SE | 1000 | 999 | 1 | 1000 | 0 |
| JE | 1000 | 942 | 58 | 1000 | 0 |
| SC | 999 | 865 | 133 | 998 | 1 |
Interpretation
Let’s see that plot again?
Interpretation
Checking residuals
Pearson's Chi-squared test
data: smoke
X-squared = 16, df = 12, p-value = 0.2Interpretation
Linguistic example
Contingency table
Code
# get corpus documents
corpus_folder <- here::here("studies", "_corpora", "clinton_trump")
fnames <- get_fnames(corpus_folder)
short_fnames <- short_names(fnames)
# get possible features
stop_list <- read_types(here::here("studies", "assets", "ca-trump-clinton", "stop_list.txt"))
features <- freqlist(fnames) %>% drop_types(stop_list) %>%
keep_pos(1:150) %>% as_types()
# create contingency table (a matrix)
d <- map(setNames(fnames, short_fnames), ~ freqlist(.x)[features]) %>%
bind_cols() %>% data.frame(row.names = features) %>%
as.matrix() %>% t() %>% drop_empty_rc()
kbl(d[1:10, 1:10]) %>% kable_paper(font_size = 20)| again | ago | also | america | american | americans | applause | are | audience | back | |
|---|---|---|---|---|---|---|---|---|---|---|
| Clinton_2016.07.28 | 4 | 5 | 6 | 24 | 3 | 11 | 113 | 36 | 6 | 6 |
| Clinton_2016.07.29 | 2 | 3 | 5 | 13 | 3 | 1 | 72 | 18 | 3 | 2 |
| Clinton_2016.08.01 | 0 | 1 | 2 | 7 | 2 | 4 | 22 | 11 | 1 | 3 |
| Clinton_2016.08.05 | 4 | 1 | 9 | 11 | 10 | 10 | 11 | 37 | 0 | 7 |
| Clinton_2016.08.10 | 0 | 0 | 5 | 6 | 4 | 3 | 41 | 19 | 0 | 4 |
| Clinton_2016.08.11 | 7 | 1 | 11 | 25 | 13 | 15 | 51 | 40 | 0 | 16 |
| Clinton_2016.08.15 | 2 | 4 | 3 | 19 | 9 | 5 | 52 | 30 | 8 | 16 |
| Clinton_2016.08.16 | 3 | 0 | 4 | 10 | 2 | 3 | 41 | 20 | 2 | 6 |
| Clinton_2016.08.17 | 3 | 2 | 5 | 12 | 6 | 12 | 64 | 34 | 0 | 2 |
| Clinton_2016.08.25 | 6 | 1 | 3 | 19 | 10 | 5 | 18 | 17 | 1 | 2 |
Linguistic example
Correspondence Analysis
Principal inertias (eigenvalues):
1 2 3 4 5 6 7
Value 0.053657 0.040702 0.022246 0.019935 0.0143 0.013147 0.011372
Percentage 14.09% 10.69% 5.84% 5.23% 3.75% 3.45% 2.99%
8 9 10 11 12 13 14
Value 0.010335 0.008489 0.007962 0.007012 0.006627 0.006017 0.005841
Percentage 2.71% 2.23% 2.09% 1.84% 1.74% 1.58% 1.53%
15 16 17 18 19 20 21
Value 0.005661 0.005539 0.00512 0.004972 0.004851 0.004673 0.004558
Percentage 1.49% 1.45% 1.34% 1.31% 1.27% 1.23% 1.2%
22 23 24 25 26 27 28
Value 0.004145 0.004037 0.003968 0.003885 0.00385 0.003568 0.003367
Percentage 1.09% 1.06% 1.04% 1.02% 1.01% 0.94% 0.88%
29 30 31 32 33 34 35
Value 0.003301 0.003141 0.003067 0.002921 0.002839 0.002717 0.002664
Percentage 0.87% 0.82% 0.81% 0.77% 0.75% 0.71% 0.7%
36 37 38 39 40 41 42
Value 0.00258 0.002534 0.002443 0.002341 0.002303 0.002214 0.002181
Percentage 0.68% 0.67% 0.64% 0.61% 0.6% 0.58% 0.57%
43 44 45 46 47 48 49
Value 0.002148 0.002031 0.001955 0.001873 0.001861 0.001794 0.001711
Percentage 0.56% 0.53% 0.51% 0.49% 0.49% 0.47% 0.45%
50 51 52 53 54 55 56
Value 0.001662 0.001653 0.001595 0.001496 0.001442 0.001406 0.001361
Percentage 0.44% 0.43% 0.42% 0.39% 0.38% 0.37% 0.36%
57 58 59 60 61 62 63
Value 0.001311 0.001267 0.001257 0.001183 0.001174 0.001111 0.001072
Percentage 0.34% 0.33% 0.33% 0.31% 0.31% 0.29% 0.28%
64 65 66 67 68 69 70
Value 0.001038 0.001011 0.00097 0.000924 0.000905 0.000842 0.000825
Percentage 0.27% 0.27% 0.25% 0.24% 0.24% 0.22% 0.22%
71 72 73 74 75 76 77
Value 0.000774 0.000768 0.000747 0.000718 0.000704 0.000671 0.000659
Percentage 0.2% 0.2% 0.2% 0.19% 0.18% 0.18% 0.17%
78 79 80 81 82 83 84
Value 0.000605 0.000595 0.000568 0.000536 0.000505 0.000481 0.00045
Percentage 0.16% 0.16% 0.15% 0.14% 0.13% 0.13% 0.12%
85 86 87 88 89 90 91
Value 0.000431 0.000422 0.000389 0.000381 0.000343 0.000335 0.000315
Percentage 0.11% 0.11% 0.1% 0.1% 0.09% 0.09% 0.08%
92 93 94 95 96 97 98
Value 0.000292 0.000275 0.000265 0.000243 0.000238 0.000217 0.000205
Percentage 0.08% 0.07% 0.07% 0.06% 0.06% 0.06% 0.05%
99 100 101 102 103 104 105
Value 0.000183 0.000173 0.000164 0.000155 0.000152 0.000148 0.000128
Percentage 0.05% 0.05% 0.04% 0.04% 0.04% 0.04% 0.03%
106 107 108 109 110 111 112 113
Value 0.000121 0.000113 8.9e-05 7.4e-05 6.8e-05 5.9e-05 5e-05 4.5e-05
Percentage 0.03% 0.03% 0.02% 0.02% 0.02% 0.02% 0.01% 0.01%
114 115 116 117
Value 3.7e-05 2.3e-05 0 0
Percentage 0.01% 0.01% 0% 0%
Rows:
Clinton_2016.07.28 Clinton_2016.07.29 Clinton_2016.08.01
Mass 0.00787 0.00516 0.00225
ChiDist 0.72854 0.65694 0.97131
Inertia 0.00418 0.00223 0.00213
Dim. 1 -2.02389 -1.16774 -1.77022
Dim. 2 0.31847 0.87857 1.00231
Clinton_2016.08.05 Clinton_2016.08.10 Clinton_2016.08.11
Mass 0.00841 0.00329 0.00770
ChiDist 0.67367 0.84070 0.85071
Inertia 0.00382 0.00233 0.00557
Dim. 1 -1.11288 -2.12166 -2.06907
Dim. 2 1.52271 1.01558 0.48095
Clinton_2016.08.15 Clinton_2016.08.16 Clinton_2016.08.17
Mass 0.01033 0.00465 0.00613
ChiDist 0.62627 0.68516 0.77281
Inertia 0.00405 0.00218 0.00366
Dim. 1 -0.61023 -1.10783 -1.82868
Dim. 2 1.32985 0.92825 1.01634
Clinton_2016.08.25 Clinton_2016.08.31 Clinton_2016.09.05.A
Mass 0.00434 0.00473 0.00946
ChiDist 0.87991 0.97231 0.62874
Inertia 0.00336 0.00448 0.00374
Dim. 1 -1.67873 -1.99464 -1.27886
Dim. 2 1.23535 0.29216 1.19872
Clinton_2016.09.05.B Clinton_2016.09.06 Clinton_2016.09.08.A
Mass 0.00517 0.00797 0.00286
ChiDist 0.71890 0.55881 1.05714
Inertia 0.00267 0.00249 0.00320
Dim. 1 -1.60773 -1.16248 -0.65555
Dim. 2 1.46085 0.75815 1.49847
Clinton_2016.09.08.B Clinton_2016.09.08.C Clinton_2016.09.29
Mass 0.00419 0.00521 0.00249
ChiDist 0.81674 0.63028 0.93480
Inertia 0.00279 0.00207 0.00217
Dim. 1 -1.81649 -1.21676 -1.38177
Dim. 2 0.81885 1.31835 1.19489
Clinton_2016.09.30 Clinton_2016.10.03 Clinton_2016.10.24
Mass 0.00509 0.00659 0.01141
ChiDist 0.78489 0.63694 0.59666
Inertia 0.00313 0.00267 0.00406
Dim. 1 -2.22845 -1.23714 -1.62939
Dim. 2 0.37197 1.19205 0.83067
Clinton_2016.10.26 Clinton_2016.10.31.A Clinton_2016.10.31.B
Mass 0.00473 0.00458 0.00462
ChiDist 0.71761 0.90953 0.74975
Inertia 0.00244 0.00379 0.00260
Dim. 1 -1.58816 -1.75994 -1.90898
Dim. 2 1.08970 1.58702 1.38698
Clinton_2016.11.01.A Clinton_2016.11.01.B Clinton_2016.11.04.A
Mass 0.00605 0.00342 0.00607
ChiDist 0.74790 1.02463 0.61072
Inertia 0.00338 0.00359 0.00226
Dim. 1 -1.29248 -2.24480 -1.38148
Dim. 2 1.37490 0.88133 0.77066
Clinton_2016.11.04.B Clinton_2016.11.05 Clinton_2016.11.06.A
Mass 0.00596 0.00130 0.00429
ChiDist 0.70280 1.13518 0.82288
Inertia 0.00294 0.00168 0.00290
Dim. 1 -1.31183 -1.94122 -2.54299
Dim. 2 1.57636 1.22924 0.89944
Clinton_2016.11.06.B Clinton_2016.11.07.A Clinton_2016.11.07.B
Mass 0.00470 0.00467 0.00580
ChiDist 0.83614 1.27873 0.68038
Inertia 0.00328 0.00763 0.00268
Dim. 1 -2.06126 -1.67631 -1.75604
Dim. 2 1.26015 1.23829 1.08815
Clinton_2016.11.07.C Clinton_2016.11.08 Clinton_2016.11.09
Mass 0.00236 0.00228 0.00291
ChiDist 1.11338 1.16716 0.96174
Inertia 0.00292 0.00310 0.00269
Dim. 1 -2.83205 -2.98628 -1.83965
Dim. 2 0.78706 0.84589 0.41832
Trump_2016.07.22 Trump_2016.07.25 Trump_2016.07.26 Trump_2016.07.27.A
Mass 0.01219 0.01690 0.00424 0.01707
ChiDist 0.69509 0.47903 0.78165 0.74738
Inertia 0.00589 0.00388 0.00259 0.00954
Dim. 1 0.69341 0.79917 -0.73954 1.16718
Dim. 2 2.27359 1.00803 -0.70112 1.73993
Trump_2016.07.27.B Trump_2016.08.01 Trump_2016.08.02 Trump_2016.08.04
Mass 0.01936 0.01814 0.02013 0.01438
ChiDist 0.45108 0.59678 0.52218 0.42709
Inertia 0.00394 0.00646 0.00549 0.00262
Dim. 1 0.97719 1.26662 1.50007 0.68727
Dim. 2 1.06519 0.76711 0.54820 0.61284
Trump_2016.08.05 Trump_2016.08.08 Trump_2016.08.10 Trump_2016.08.12.A
Mass 0.01671 0.00668 0.01773 0.01699
ChiDist 0.44280 1.07661 0.53492 0.44378
Inertia 0.00328 0.00774 0.00507 0.00335
Dim. 1 0.64611 -1.54888 1.45535 1.10950
Dim. 2 0.69654 -2.21025 0.88880 0.70871
Trump_2016.08.12.B Trump_2016.08.15 Trump_2016.08.16 Trump_2016.08.17
Mass 0.01563 0.00589 0.00616 0.01222
ChiDist 0.48574 1.00223 0.68532 0.67146
Inertia 0.00369 0.00592 0.00290 0.00551
Dim. 1 1.28463 -1.24694 -1.13363 0.48849
Dim. 2 0.17365 -0.14256 -1.57701 1.19286
Trump_2016.08.18 Trump_2016.08.19 Trump_2016.08.22 Trump_2016.08.23
Mass 0.00646 0.00728 0.00785 0.00729
ChiDist 0.71089 0.67168 0.51183 0.91938
Inertia 0.00327 0.00329 0.00206 0.00616
Dim. 1 -1.47111 -0.67781 -0.43326 -0.80832
Dim. 2 -1.75252 -1.90305 -1.24625 -1.99774
Trump_2016.08.24.A Trump_2016.08.24.B Trump_2016.08.25 Trump_2016.08.30
Mass 0.01321 0.00721 0.00782 0.00732
ChiDist 0.45717 0.70808 0.61256 0.54672
Inertia 0.00276 0.00362 0.00293 0.00219
Dim. 1 0.30969 -1.06200 -0.35949 -0.48429
Dim. 2 -1.39983 -1.93394 -1.48243 -0.99443
Trump_2016.08.31 Trump_2016.09.01.A Trump_2016.09.01.B
Mass 0.01058 0.00495 0.00262
ChiDist 0.52025 0.75350 1.00512
Inertia 0.00286 0.00281 0.00265
Dim. 1 -0.31778 -0.33167 -1.74930
Dim. 2 -0.68690 -2.01326 -1.75483
Trump_2016.09.06.A Trump_2016.09.06.B Trump_2016.09.07.A
Mass 0.01347 0.00939 0.00451
ChiDist 0.56006 0.43420 0.90409
Inertia 0.00422 0.00177 0.00369
Dim. 1 0.98309 0.06941 -0.93760
Dim. 2 1.07593 -0.96084 -1.17754
Trump_2016.09.07.B Trump_2016.09.08 Trump_2016.09.09.A
Mass 0.00625 0.00616 0.00934
ChiDist 0.78935 0.67009 0.46204
Inertia 0.00390 0.00277 0.00199
Dim. 1 0.81699 0.02700 0.37344
Dim. 2 1.15679 -0.13474 -0.03671
Trump_2016.09.09.B Trump_2016.09.09 Trump_2016.09.12.A
Mass 0.00329 0.00934 0.00912
ChiDist 0.76311 0.46204 0.51056
Inertia 0.00192 0.00199 0.00238
Dim. 1 0.34694 0.37344 -0.19505
Dim. 2 -1.12001 -0.03671 -0.73983
Trump_2016.09.12.B Trump_2016.09.13.A Trump_2016.09.13.B
Mass 0.00309 0.00688 0.00314
ChiDist 0.97906 0.58271 1.62463
Inertia 0.00296 0.00233 0.00830
Dim. 1 -1.14776 -0.14949 -1.89710
Dim. 2 -1.07806 -1.45653 -0.40451
Trump_2016.09.13 Trump_2016.09.16 Trump_2016.09.17 Trump_2016.09.29
Mass 0.00688 0.00957 0.00444 0.00790
ChiDist 0.58271 0.48383 0.79010 0.52557
Inertia 0.00233 0.00224 0.00277 0.00218
Dim. 1 -0.14949 0.25199 -0.73154 0.20163
Dim. 2 -1.45653 -0.73164 0.06891 -0.95380
Trump_2016.09.30 Trump_2016.10.03.A Trump_2016.10.03.B Trump_2016.10.05
Mass 0.01137 0.01270 0.01131 0.01167
ChiDist 0.53608 0.69566 0.46560 0.37763
Inertia 0.00327 0.00614 0.00245 0.00166
Dim. 1 0.07976 0.54541 -0.16163 -0.07123
Dim. 2 -1.11114 0.82933 -0.54760 -0.67281
Trump_2016.10.06 Trump_2016.10.10.A Trump_2016.10.10.B Trump_2016.10.11
Mass 0.01621 0.01286 0.01505 0.01321
ChiDist 0.46661 0.46752 0.41148 0.39338
Inertia 0.00353 0.00281 0.00255 0.00204
Dim. 1 0.92687 0.67948 0.93722 0.37293
Dim. 2 1.07343 -0.38991 -0.18854 -0.30168
Trump_2016.10.12 Trump_2016.10.13 Trump_2016.10.14.A Trump_2016.10.14.B
Mass 0.0142 0.00725 0.00986 0.01447
ChiDist 0.4966 0.63648 0.44339 0.40301
Inertia 0.0035 0.00294 0.00194 0.00235
Dim. 1 0.6583 -0.45674 0.81191 0.68835
Dim. 2 -0.0311 -0.48487 -0.35958 -0.13796
Trump_2016.10.17 Trump_2016.10.18 Trump_2016.10.21.A Trump_2016.10.21.B
Mass 0.00999 0.00936 0.00600 0.01024
ChiDist 0.75894 0.48127 0.75496 0.51578
Inertia 0.00575 0.00217 0.00342 0.00272
Dim. 1 0.28669 0.08980 1.10636 0.52937
Dim. 2 -0.68317 -0.71455 0.92033 -0.87859
Trump_2016.10.21.C Trump_2016.10.27 Trump_2016.10.28 Trump_2016.10.31
Mass 0.00641 0.00722 0.00986 0.01258
ChiDist 0.60011 0.53977 0.50616 0.45557
Inertia 0.00231 0.00210 0.00253 0.00261
Dim. 1 -0.14318 0.22554 0.55159 0.25615
Dim. 2 -1.44395 -0.87037 -0.37745 -0.85616
Trump_2016.11.01.A Trump_2016.11.01.B Trump_2016.11.02.A
Mass 0.00875 0.00403 0.00888
ChiDist 0.47734 0.81191 0.57400
Inertia 0.00199 0.00266 0.00293
Dim. 1 -0.21015 -0.63058 0.24179
Dim. 2 -1.17654 -0.96360 -1.31010
Trump_2016.11.02.B Trump_2016.11.02.C Trump_2016.11.03.A
Mass 0.01078 0.00696 0.01122
ChiDist 0.44223 0.45933 0.48062
Inertia 0.00211 0.00147 0.00259
Dim. 1 0.43943 0.25498 0.33105
Dim. 2 -0.38575 -0.71532 -0.89609
Trump_2016.11.03.B Trump_2016.11.04.A Trump_2016.11.04.B
Mass 0.01030 0.01010 0.00943
ChiDist 0.40904 0.50267 0.42586
Inertia 0.00172 0.00255 0.00171
Dim. 1 0.59619 0.41608 0.23570
Dim. 2 -0.74128 -0.41939 -0.49207
Trump_2016.11.04.C Trump_2016.11.05.A Trump_2016.11.05.B
Mass 0.01235 0.01250 0.00973
ChiDist 0.46700 0.44782 0.53837
Inertia 0.00269 0.00251 0.00282
Dim. 1 0.86480 0.81109 0.00957
Dim. 2 -0.35621 -0.44001 -0.72408
Trump_2016.11.06 Trump_2016.11.07.A Trump_2016.11.07.B
Mass 0.00769 0.0115 0.01174
ChiDist 0.54634 0.5817 0.45662
Inertia 0.00230 0.0039 0.00245
Dim. 1 0.55748 0.4638 0.69168
Dim. 2 -0.71745 -0.5945 -0.50859
Trump_2016.11.07.C Trump_2016.11.07.D Trump_2016.11.08 Trump_2016.11.09
Mass 0.00684 0.00934 0.00819 0.00307
ChiDist 0.70994 0.46062 0.54275 0.95258
Inertia 0.00345 0.00198 0.00241 0.00279
Dim. 1 0.32096 -0.11529 0.12273 -0.83664
Dim. 2 -1.13683 -0.71337 -1.09737 0.50210
Columns:
again ago also america american americans applause
Mass 0.00520 0.00239 0.00441 0.00728 0.00623 0.00298 0.0417
ChiDist 0.59271 0.67304 0.69401 0.74844 1.00224 1.01361 0.5077
Inertia 0.00183 0.00108 0.00213 0.00407 0.00626 0.00306 0.0107
Dim. 1 -0.04653 1.42091 -1.32777 -2.51700 -1.95816 -3.08274 -1.3551
Dim. 2 -1.91206 -0.14646 -1.12235 -0.39190 -2.55831 -0.80550 -0.8807
are audience back bad been being believe
Mass 0.02665 0.00398 0.00789 0.00337 0.00913 0.002714 0.00574
ChiDist 0.27224 1.31493 0.45793 0.74207 0.40025 0.580570 0.66284
Inertia 0.00198 0.00688 0.00165 0.00186 0.00146 0.000915 0.00252
Dim. 1 -0.29785 -0.82076 0.16778 1.97468 0.01503 -0.143460 -1.34014
Dim. 2 -0.33262 -3.10009 -0.65682 -0.55942 -0.16116 -0.042899 0.28557
better big booing bring build campaign can't care
Mass 0.00333 0.004299 0.00398 0.00304 0.00310 0.00319 0.00413 0.00393
ChiDist 0.56943 0.455725 1.32978 0.70156 0.67447 0.94575 0.57161 0.92892
Inertia 0.00108 0.000893 0.00704 0.00150 0.00141 0.00286 0.00135 0.00339
Dim. 1 -0.32014 0.674739 0.13660 -0.25543 0.30632 -2.23064 0.45155 -0.40253
Dim. 2 -0.03827 -0.033457 -3.15327 -1.65955 -0.88945 0.53180 0.57872 -0.24827
change clinton come countries country day deal did
Mass 0.00265 0.00995 0.00421 0.00244 0.01409 0.00310 0.00289 0.00494
ChiDist 1.03829 0.66937 0.57192 0.70936 0.41172 0.68683 0.79254 0.56255
Inertia 0.00286 0.00446 0.00138 0.00123 0.00239 0.00146 0.00181 0.00156
Dim. 1 -0.63910 -0.95596 0.46960 0.68959 -0.45816 -0.69089 1.22428 0.55516
Dim. 2 -1.89876 -0.76519 0.11646 -1.12318 -1.28842 0.97911 -0.84514 1.17934
didn't doesn't doing don't donald done election even
Mass 0.00293 0.00272 0.00434 0.01298 0.00429 0.00281 0.00227 0.00488
ChiDist 0.70625 0.70162 0.53789 0.42538 1.01695 0.76950 1.33125 0.49803
Inertia 0.00146 0.00134 0.00126 0.00235 0.00444 0.00166 0.00401 0.00121
Dim. 1 1.03604 -0.09742 1.01116 1.21532 -2.10531 -0.75504 -3.41121 0.19051
Dim. 2 1.08141 0.65338 0.25113 0.45578 2.23348 0.86540 2.17855 0.39306
ever everybody folks four get give go going
Mass 0.00465 0.00344 0.00403 0.00234 0.01169 0.00271 0.00637 0.03682
ChiDist 0.52537 0.65586 0.68399 0.74628 0.34888 0.74148 0.54306 0.39574
Inertia 0.00128 0.00148 0.00189 0.00130 0.00142 0.00149 0.00188 0.00577
Dim. 1 0.21894 -0.65236 1.49939 0.75926 -0.18967 -0.18898 0.08363 1.06378
Dim. 2 -1.24393 0.57027 -1.15309 -1.40534 0.51270 0.90447 1.38814 -0.63649
gonna good got government great had happen has
Mass 0.00261 0.00590 0.00708 0.00238 0.01249 0.00674 0.00326 0.00903
ChiDist 1.97901 0.50567 0.61873 0.89879 0.42992 0.59454 0.65432 0.53272
Inertia 0.01024 0.00151 0.00271 0.00193 0.00231 0.00238 0.00140 0.00256
Dim. 1 1.74616 0.20730 -0.22118 -0.38390 0.75507 0.11124 0.97980 -1.05348
Dim. 2 -0.72854 0.99638 1.39320 -2.75894 -0.27813 1.57321 -1.20735 0.18450
he's help hillary i'll i'm i've important inaudible
Mass 0.00369 0.00231 0.01102 0.00237 0.00949 0.00441 0.00238 0.00301
ChiDist 0.87298 1.14192 0.53795 0.67733 0.43234 0.64769 0.71351 2.50099
Inertia 0.00281 0.00301 0.00319 0.00109 0.00177 0.00185 0.00121 0.01881
Dim. 1 0.52917 -3.53872 0.00955 0.83712 0.32248 -0.23131 -0.99611 0.12751
Dim. 2 2.59097 1.73090 -1.46143 0.63621 0.35250 1.52470 0.56070 1.38804
is isis it's job jobs just keep know
Mass 0.03485 0.00246 0.01963 0.00301 0.00926 0.01176 0.00237 0.01936
ChiDist 0.28776 1.37982 0.33181 0.79021 0.67097 0.34990 0.81525 0.37735
Inertia 0.00288 0.00468 0.00216 0.00188 0.00417 0.00144 0.00158 0.00276
Dim. 1 -0.70550 0.56684 0.92092 -0.56673 -0.33775 -0.19619 -0.96505 0.32623
Dim. 2 0.24544 -0.23310 -0.11132 0.08972 -2.12776 0.87747 -0.18902 1.41148
laughter let like look lot love made make
Mass 0.00235 0.00334 0.01013 0.00658 0.00591 0.00374 0.00279 0.00860
ChiDist 1.07469 0.61233 0.34748 0.62405 0.52565 0.70683 0.68018 0.47955
Inertia 0.00271 0.00125 0.00122 0.00256 0.00163 0.00187 0.00129 0.00198
Dim. 1 0.12941 0.07260 0.78828 1.24481 0.80809 0.52054 -0.15455 -1.25143
Dim. 2 2.16122 0.13605 0.15329 0.70219 1.22086 -0.01538 -0.06180 0.16861
mean mexico military million money need new not
Mass 0.00355 0.00230 0.00245 0.00249 0.00398 0.00362 0.00638 0.01767
ChiDist 0.82322 0.94229 1.35259 0.85416 0.61222 0.78488 0.83331 0.29258
Inertia 0.00241 0.00204 0.00449 0.00182 0.00149 0.00223 0.00443 0.00151
Dim. 1 1.69345 1.91787 0.18755 0.14550 1.39654 -1.44421 -0.97159 -0.60096
Dim. 2 1.76899 -1.36846 -0.48560 0.32542 -0.44203 1.00628 -0.92025 0.43422
obama ok only out pay people percent ph
Mass 0.00288 0.00363 0.00326 0.0116 0.00308 0.02483 0.00455 0.00232
ChiDist 0.82744 0.92138 0.73986 0.3476 0.86535 0.27217 0.70893 1.29998
Inertia 0.00197 0.00308 0.00179 0.0014 0.00231 0.00184 0.00229 0.00392
Dim. 1 0.38621 2.77864 -1.13106 -0.1124 -1.02251 0.07947 0.88837 1.52291
Dim. 2 -0.33202 1.50829 -1.05026 0.4996 1.00648 0.09766 -2.07962 1.61024
place plan president put really remember right said
Mass 0.00227 0.00237 0.00591 0.003444 0.00565 0.00318 0.01144 0.01234
ChiDist 0.69791 1.07109 0.80612 0.521433 0.61536 0.71638 0.37559 0.56708
Inertia 0.00110 0.00272 0.00384 0.000936 0.00214 0.00163 0.00161 0.00397
Dim. 1 -0.05062 -1.55305 -2.08299 -0.392729 -0.55240 0.28864 0.86696 1.18909
Dim. 2 -0.23084 -1.41184 1.70011 -0.149998 1.89038 -1.08873 0.17230 1.61726
say see seen she she's state states stop
Mass 0.00847 0.00582 0.00238 0.01542 0.00367 0.00411 0.00449 0.00235
ChiDist 0.37242 0.50491 0.76344 0.57268 0.73651 0.68547 0.49413 0.78727
Inertia 0.00118 0.00148 0.00138 0.00506 0.00199 0.00193 0.00110 0.00145
Dim. 1 0.60954 0.68700 0.37797 0.90144 1.54675 -0.61681 -0.09087 0.69153
Dim. 2 0.65385 0.91515 -0.17770 -1.18574 -0.69465 -1.31118 -0.51880 -1.66137
take talk tax tell thank that's there's these
Mass 0.00571 0.00238 0.00225 0.00549 0.00881 0.01031 0.00228 0.00734
ChiDist 0.46795 0.83659 1.36466 0.47710 0.75745 0.37205 0.73098 0.51526
Inertia 0.00125 0.00167 0.00418 0.00125 0.00505 0.00143 0.00122 0.00195
Dim. 1 0.54986 -0.31465 -1.08180 -0.05398 -1.08654 0.19579 0.28661 0.67819
Dim. 2 0.00448 1.34102 -0.33456 0.55411 -0.92050 0.48380 0.54697 -0.42013
they're thing things think those time together too
Mass 0.00977 0.00321 0.003849 0.00900 0.00454 0.00660 0.00307 0.00231
ChiDist 0.62821 0.65050 0.503108 0.51260 0.57317 0.45555 1.21279 0.78833
Inertia 0.00386 0.00136 0.000974 0.00237 0.00149 0.00137 0.00451 0.00144
Dim. 1 2.14070 1.26334 0.559702 0.42815 -1.18586 -0.29843 -4.07737 -0.99743
Dim. 2 -0.06408 1.27331 0.620920 1.61146 0.23093 -0.24678 1.51046 0.36597
trade trump united up very vote wall want
Mass 0.00346 0.01794 0.00379 0.01013 0.01174 0.00471 0.00311 0.01230
ChiDist 0.94017 0.59361 0.60364 0.35919 0.54847 0.89731 0.86193 0.47207
Inertia 0.00305 0.00632 0.00138 0.00131 0.00353 0.00379 0.00231 0.00274
Dim. 1 1.18794 0.32062 -0.44590 -0.31289 0.78273 -1.42633 1.06967 -0.65314
Dim. 2 -1.90379 -0.04606 -0.91234 0.59942 -0.35590 -0.48455 -2.04138 0.88367
wants was way we'll we're well were what's
Mass 0.00283 0.01763 0.00633 0.00258 0.02226 0.00521 0.00554 0.00357
ChiDist 0.64370 0.47071 0.37498 0.81727 0.44595 0.65637 0.63980 0.59124
Inertia 0.00117 0.00391 0.00089 0.00172 0.00443 0.00224 0.00227 0.00125
Dim. 1 0.55184 0.35683 0.56292 0.86412 1.07606 -0.15334 -0.32015 1.46406
Dim. 2 -0.77502 1.40292 0.12552 -1.16637 -0.54869 2.18059 0.97466 -0.12975
why win work working world years you're your
Mass 0.00339 0.00437 0.00482 0.00265 0.00444 0.00598 0.00457 0.01139
ChiDist 0.65960 0.84503 1.02405 0.89373 0.52063 0.45065 0.60997 0.47686
Inertia 0.00147 0.00312 0.00505 0.00212 0.00120 0.00121 0.00170 0.00259
Dim. 1 -0.47816 1.28319 -3.13882 -2.07193 -0.36458 0.39163 1.03397 -0.91277
Dim. 2 0.74921 -1.31044 1.11788 0.40909 -0.48495 -0.53551 0.52202 -0.65435
Principal inertias (eigenvalues):
dim value % cum% scree plot
1 0.053657 14.1 14.1 ****
2 0.040702 10.7 24.8 ***
3 0.022246 5.8 30.6 *
4 0.019935 5.2 35.8 *
5 0.014300 3.8 39.6 *
6 0.013147 3.5 43.1 *
7 0.011372 3.0 46.0 *
8 0.010335 2.7 48.8 *
9 0.008489 2.2 51.0 *
10 0.007962 2.1 53.1 *
11 0.007012 1.8 54.9
12 0.006627 1.7 56.7
13 0.006017 1.6 58.2
14 0.005841 1.5 59.8
15 0.005661 1.5 61.3
16 0.005539 1.5 62.7
17 0.005120 1.3 64.1
18 0.004972 1.3 65.4
19 0.004851 1.3 66.6
20 0.004673 1.2 67.9
21 0.004558 1.2 69.1
22 0.004145 1.1 70.1
23 0.004037 1.1 71.2
24 0.003968 1.0 72.2
25 0.003885 1.0 73.3
26 0.003850 1.0 74.3
27 0.003568 0.9 75.2
28 0.003367 0.9 76.1
29 0.003301 0.9 77.0
30 0.003141 0.8 77.8
31 0.003067 0.8 78.6
32 0.002921 0.8 79.4
33 0.002839 0.7 80.1
34 0.002717 0.7 80.8
35 0.002664 0.7 81.5
36 0.002580 0.7 82.2
37 0.002534 0.7 82.9
38 0.002443 0.6 83.5
39 0.002341 0.6 84.1
40 0.002303 0.6 84.7
41 0.002214 0.6 85.3
42 0.002181 0.6 85.9
43 0.002148 0.6 86.4
44 0.002031 0.5 87.0
45 0.001955 0.5 87.5
46 0.001873 0.5 88.0
47 0.001861 0.5 88.5
48 0.001794 0.5 88.9
49 0.001711 0.4 89.4
50 0.001662 0.4 89.8
51 0.001653 0.4 90.3
52 0.001595 0.4 90.7
53 0.001496 0.4 91.1
54 0.001442 0.4 91.4
55 0.001406 0.4 91.8
56 0.001361 0.4 92.2
57 0.001311 0.3 92.5
58 0.001267 0.3 92.9
59 0.001257 0.3 93.2
60 0.001183 0.3 93.5
61 0.001174 0.3 93.8
62 0.001111 0.3 94.1
63 0.001072 0.3 94.4
64 0.001038 0.3 94.6
65 0.001011 0.3 94.9
66 0.000970 0.3 95.2
67 0.000924 0.2 95.4
68 0.000905 0.2 95.6
69 0.000842 0.2 95.9
70 0.000825 0.2 96.1
71 0.000774 0.2 96.3
72 0.000768 0.2 96.5
73 0.000747 0.2 96.7
74 0.000718 0.2 96.9
75 0.000704 0.2 97.1
76 0.000671 0.2 97.2
77 0.000659 0.2 97.4
78 0.000605 0.2 97.6
79 0.000595 0.2 97.7
80 0.000568 0.1 97.9
81 0.000536 0.1 98.0
82 0.000505 0.1 98.1
83 0.000481 0.1 98.3
84 0.000450 0.1 98.4
85 0.000431 0.1 98.5
86 0.000422 0.1 98.6
87 0.000389 0.1 98.7
88 0.000381 0.1 98.8
89 0.000343 0.1 98.9
90 0.000335 0.1 99.0
91 0.000315 0.1 99.1
92 0.000292 0.1 99.2
93 0.000275 0.1 99.2
94 0.000265 0.1 99.3
95 0.000243 0.1 99.4
96 0.000238 0.1 99.4
97 0.000217 0.1 99.5
98 0.000205 0.1 99.5
99 0.000183 0.0 99.6
100 0.000173 0.0 99.6
101 0.000164 0.0 99.7
102 0.000155 0.0 99.7
103 0.000152 0.0 99.7
104 0.000148 0.0 99.8
105 0.000128 0.0 99.8
106 0.000121 0.0 99.9
107 0.000113 0.0 99.9
108 8.9e-050 0.0 99.9
109 7.4e-050 0.0 99.9
110 6.8e-050 0.0 99.9
111 5.9e-050 0.0 100.0
112 5e-05000 0.0 100.0
113 4.5e-050 0.0 100.0
114 3.7e-050 0.0 100.0
115 2.3e-050 0.0 100.0
116 00000000 0.0 100.0
117 00000000 0.0 100.0
-------- -----
Total: 0.380871 100.0
Rows:
name mass qlt inr k=1 cor ctr k=2 cor ctr
1 | C_20160728 | 8 422 11 | -469 414 32 | 64 8 1 |
2 | C_20160729 | 5 242 6 | -270 170 7 | 177 73 4 |
3 | C_20160801 | 2 222 6 | -410 178 7 | 202 43 2 |
4 | C_20160805 | 8 354 10 | -258 146 10 | 307 208 20 |
5 | C_20160810 | 3 401 6 | -491 342 15 | 205 59 3 |
6 | C_20160811 | 8 330 15 | -479 317 33 | 97 13 2 |
7 | C_20160815 | 10 234 11 | -141 51 4 | 268 184 18 |
8 | C_20160816 | 5 215 6 | -257 140 6 | 187 75 4 |
9 | C_20160817 | 6 371 10 | -424 300 21 | 205 70 6 |
10 | C_2016082 | 4 276 9 | -389 195 12 | 249 80 7 |
11 | C_2016083 | 5 229 12 | -462 226 19 | 59 4 0 |
12 | C_20160905A | 9 370 10 | -296 222 15 | 242 148 14 |
13 | C_20160905B | 5 436 7 | -372 268 13 | 295 168 11 |
14 | C_20160906 | 8 307 7 | -269 232 11 | 153 75 5 |
15 | C_20160908A | 3 102 8 | -152 21 1 | 302 82 6 |
16 | C_20160908B | 4 306 7 | -421 265 14 | 165 41 3 |
17 | C_20160908C | 5 378 5 | -282 200 8 | 266 178 9 |
18 | C_2016092 | 2 184 6 | -320 117 5 | 241 67 4 |
19 | C_2016093 | 5 442 8 | -516 433 25 | 75 9 1 |
20 | C_2016100 | 7 345 7 | -287 202 10 | 240 143 9 |
21 | C_20161024 | 11 479 11 | -377 400 30 | 168 79 8 |
22 | C_20161026 | 5 357 6 | -368 263 12 | 220 94 6 |
23 | C_20161031A | 5 325 10 | -408 201 14 | 320 124 12 |
24 | C_20161031B | 5 487 7 | -442 348 17 | 280 139 9 |
25 | C_20161101A | 6 298 9 | -299 160 10 | 277 138 11 |
26 | C_20161101B | 3 288 9 | -520 258 17 | 178 30 3 |
27 | C_20161104A | 6 339 6 | -320 275 12 | 155 65 4 |
28 | C_20161104B | 6 392 8 | -304 187 10 | 318 205 15 |
29 | C_20161105 | 1 205 4 | -450 157 5 | 248 48 2 |
30 | C_20161106A | 4 561 8 | -589 512 28 | 181 49 3 |
31 | C_20161106B | 5 419 9 | -477 326 20 | 254 92 7 |
32 | C_20161107A | 5 130 20 | -388 92 13 | 250 38 7 |
33 | C_20161107B | 6 462 7 | -407 357 18 | 220 104 7 |
34 | C_20161107C | 2 368 8 | -656 347 19 | 159 20 1 |
35 | C_20161108 | 2 373 8 | -692 351 20 | 171 21 2 |
36 | C_20161109 | 3 204 7 | -426 196 10 | 84 8 1 |
37 | T_20160722 | 12 489 15 | 161 53 6 | 459 435 63 |
38 | T_20160725 | 17 330 10 | 185 149 11 | 203 180 17 |
39 | T_20160726 | 4 81 7 | -171 48 2 | -141 33 2 |
40 | T_20160727A | 17 351 25 | 270 131 23 | 351 221 52 |
41 | T_20160727B | 19 479 10 | 226 252 18 | 215 227 22 |
42 | T_20160801 | 18 309 17 | 293 242 29 | 155 67 11 |
43 | T_20160802 | 20 488 14 | 347 443 45 | 111 45 6 |
44 | T_20160804 | 14 223 7 | 159 139 7 | 124 84 5 |
45 | T_20160805 | 17 215 9 | 150 114 7 | 141 101 8 |
46 | T_20160808 | 7 283 20 | -359 111 16 | -446 172 33 |
47 | T_20160810 | 18 510 13 | 337 397 38 | 179 112 14 |
48 | T_20160812A | 17 439 9 | 257 335 21 | 143 104 9 |
49 | T_20160812B | 16 380 10 | 298 375 26 | 35 5 0 |
50 | T_20160815 | 6 84 16 | -289 83 9 | -29 1 0 |
51 | T_20160816 | 6 362 8 | -263 147 8 | -318 216 15 |
52 | T_20160817 | 12 157 14 | 113 28 3 | 241 128 17 |
53 | T_20160818 | 6 477 9 | -341 230 14 | -354 247 20 |
54 | T_20160819 | 7 381 9 | -157 55 3 | -384 327 26 |
55 | T_20160822 | 8 280 5 | -100 38 1 | -251 241 12 |
56 | T_20160823 | 7 234 16 | -187 41 5 | -403 192 29 |
57 | T_20160824A | 13 406 7 | 72 25 1 | -282 382 26 |
58 | T_20160824B | 7 424 9 | -246 121 8 | -390 304 27 |
59 | T_20160825 | 8 257 8 | -83 18 1 | -299 238 17 |
60 | T_20160830 | 7 177 6 | -112 42 2 | -201 135 7 |
61 | T_20160831 | 11 91 8 | -74 20 1 | -139 71 5 |
62 | T_20160901A | 5 301 7 | -77 10 1 | -406 291 20 |
63 | T_20160901B | 3 287 7 | -405 163 8 | -354 124 8 |
64 | T_20160906A | 13 316 11 | 228 165 13 | 217 150 16 |
65 | T_20160906B | 9 201 5 | 16 1 0 | -194 199 9 |
66 | T_20160907A | 5 127 10 | -217 58 4 | -238 69 6 |
67 | T_20160907B | 6 145 10 | 189 57 4 | 233 87 8 |
68 | T_20160908 | 6 2 7 | 6 0 0 | -27 2 0 |
69 | T_20160909A | 9 35 5 | 87 35 1 | -7 0 0 |
70 | T_20160909B | 3 99 5 | 80 11 0 | -226 88 4 |
71 | Tr_20160909 | 9 35 5 | 87 35 1 | -7 0 0 |
72 | T_20160912A | 9 93 6 | -45 8 0 | -149 85 5 |
73 | T_20160912B | 3 123 8 | -266 74 4 | -217 49 4 |
74 | T_20160913A | 7 258 6 | -35 4 0 | -294 254 15 |
75 | T_20160913B | 3 76 22 | -439 73 11 | -82 3 1 |
76 | Tr_20160913 | 7 258 6 | -35 4 0 | -294 254 15 |
77 | T_20160916 | 10 108 6 | 58 15 1 | -148 93 5 |
78 | T_20160917 | 4 46 7 | -169 46 2 | 14 0 0 |
79 | T_2016092 | 8 142 6 | 47 8 0 | -192 134 7 |
80 | T_2016093 | 11 176 9 | 18 1 0 | -224 175 14 |
81 | T_20161003A | 13 91 16 | 126 33 4 | 167 58 9 |
82 | T_20161003B | 11 63 6 | -37 6 0 | -110 56 3 |
83 | T_20161005 | 12 131 4 | -16 2 0 | -136 129 5 |
84 | T_20161006 | 16 427 9 | 215 212 14 | 217 215 19 |
85 | T_20161010A | 13 142 7 | 157 113 6 | -79 28 2 |
86 | T_20161010B | 15 287 7 | 217 278 13 | -38 9 1 |
87 | T_20161011 | 13 72 5 | 86 48 2 | -61 24 1 |
88 | T_20161012 | 14 94 9 | 152 94 6 | -6 0 0 |
89 | T_20161013 | 7 51 8 | -106 28 2 | -98 24 2 |
90 | T_20161014A | 10 207 5 | 188 180 6 | -73 27 1 |
91 | T_20161014B | 14 161 6 | 159 157 7 | -28 5 0 |
92 | T_20161017 | 10 41 15 | 66 8 1 | -138 33 5 |
93 | T_20161018 | 9 92 6 | 21 2 0 | -144 90 5 |
94 | T_20161021A | 6 176 9 | 256 115 7 | 186 60 5 |
95 | T_20161021B | 10 175 7 | 123 57 3 | -177 118 8 |
96 | T_20161021C | 6 239 6 | -33 3 0 | -291 236 13 |
97 | T_20161027 | 7 115 6 | 52 9 0 | -176 106 5 |
98 | T_20161028 | 10 86 7 | 128 64 3 | -76 23 1 |
99 | T_2016103 | 13 161 7 | 59 17 1 | -173 144 9 |
100 | T_20161101A | 9 258 5 | -49 10 0 | -237 247 12 |
101 | T_20161101B | 4 90 7 | -146 32 2 | -194 57 4 |
102 | T_20161102A | 9 222 8 | 56 10 1 | -264 212 15 |
103 | T_20161102B | 11 84 6 | 102 53 2 | -78 31 2 |
104 | T_20161102C | 7 115 4 | 59 17 0 | -144 99 4 |
105 | T_20161103A | 11 167 7 | 77 25 1 | -181 141 9 |
106 | T_20161103B | 10 248 5 | 138 114 4 | -150 134 6 |
107 | T_20161104A | 10 65 7 | 96 37 2 | -85 28 2 |
108 | T_20161104B | 9 71 4 | 55 16 1 | -99 54 2 |
109 | T_20161104C | 12 208 7 | 200 184 9 | -72 24 2 |
110 | T_20161105A | 12 215 7 | 188 176 8 | -89 39 2 |
111 | T_20161105B | 10 74 7 | 2 0 0 | -146 74 5 |
112 | T_20161106 | 8 126 6 | 129 56 2 | -145 70 4 |
113 | T_20161107A | 12 77 10 | 107 34 2 | -120 43 4 |
114 | T_20161107B | 12 174 6 | 160 123 6 | -103 50 3 |
115 | T_20161107C | 7 115 9 | 74 11 1 | -229 104 9 |
116 | T_20161107D | 9 101 5 | -27 3 0 | -144 98 5 |
117 | T_20161108 | 8 169 6 | 28 3 0 | -221 166 10 |
118 | T_20161109 | 3 53 7 | -194 41 2 | 101 11 1 |
Columns:
name mass qlt inr k=1 cor ctr k=2 cor ctr
1 | agan | 5 424 5 | -11 0 0 | -386 424 19 |
2 | ago | 2 241 3 | 329 239 5 | -30 2 0 |
3 | also | 4 303 6 | -308 196 8 | -226 106 6 |
4 | amerc | 7 618 11 | -583 607 46 | -79 11 1 |
5 | amercn | 6 470 16 | -454 205 24 | -516 265 41 |
6 | amrcns | 3 522 8 | -714 496 28 | -163 26 2 |
7 | appl | 42 505 28 | -314 382 77 | -178 122 32 |
8 | are | 27 125 5 | -69 64 2 | -67 61 3 |
9 | adnc | 4 247 18 | -190 21 3 | -625 226 38 |
10 | back | 8 91 4 | 39 7 0 | -133 84 3 |
11 | bad | 3 403 5 | 457 380 13 | -113 23 1 |
12 | been | 9 7 4 | 3 0 0 | -33 7 0 |
13 | beng | 3 3 2 | -33 3 0 | -9 0 0 |
14 | belv | 6 227 7 | -310 219 10 | 58 8 0 |
15 | bttr | 3 17 3 | -74 17 0 | -8 0 0 |
16 | big | 4 118 2 | 156 118 2 | -7 0 0 |
17 | bong | 4 229 18 | 32 1 0 | -636 229 40 |
18 | brng | 3 235 4 | -59 7 0 | -335 228 8 |
19 | buld | 3 82 4 | 71 11 0 | -179 71 2 |
20 | cmpg | 3 311 8 | -517 298 16 | 107 13 1 |
21 | cant | 4 75 4 | 105 33 1 | 117 42 1 |
22 | care | 4 13 9 | -93 10 1 | -50 3 0 |
23 | chng | 3 156 8 | -148 20 1 | -383 136 10 |
24 | clnt | 10 163 12 | -221 109 9 | -154 53 6 |
25 | come | 4 38 4 | 109 36 1 | 23 2 0 |
26 | cntrs | 2 153 3 | 160 51 1 | -227 102 3 |
27 | cntry | 14 465 6 | -106 66 3 | -260 399 23 |
28 | day | 3 137 4 | -160 54 1 | 198 83 3 |
29 | deal | 3 174 5 | 284 128 4 | -171 46 2 |
30 | did | 5 231 4 | 129 52 2 | 238 179 7 |
31 | ddnt | 3 211 4 | 240 115 3 | 218 95 3 |
32 | dsnt | 3 36 4 | -23 1 0 | 132 35 1 |
33 | dong | 4 198 3 | 234 190 4 | 51 9 0 |
34 | dont | 13 485 6 | 282 438 19 | 92 47 3 |
35 | dnld | 4 426 12 | -488 230 19 | 451 196 21 |
36 | done | 3 103 4 | -175 52 2 | 175 51 2 |
37 | elct | 2 461 11 | -790 352 26 | 440 109 11 |
38 | even | 5 33 3 | 44 8 0 | 79 25 1 |
39 | ever | 5 237 3 | 51 9 0 | -251 228 7 |
40 | evry | 3 84 4 | -151 53 1 | 115 31 1 |
41 | flks | 4 374 5 | 347 258 9 | -233 116 5 |
42 | four | 2 200 3 | 176 56 1 | -284 144 5 |
43 | get | 12 104 4 | -44 16 0 | 103 88 3 |
44 | give | 3 64 4 | -44 3 0 | 182 61 2 |
45 | go | 6 267 5 | 19 1 0 | 280 266 12 |
46 | gong | 37 493 15 | 246 388 42 | -128 105 15 |
47 | gonn | 3 47 27 | 404 42 8 | -147 6 1 |
48 | good | 6 167 4 | 48 9 0 | 201 158 6 |
49 | got | 7 213 7 | -51 7 0 | 281 206 14 |
50 | gvrn | 2 393 5 | -89 10 0 | -557 384 18 |
51 | gret | 12 183 6 | 175 166 7 | -56 17 1 |
52 | had | 7 287 6 | 26 2 0 | 317 285 17 |
53 | hppn | 3 259 4 | 227 120 3 | -244 139 5 |
54 | has | 9 215 7 | -244 210 10 | 37 5 0 |
55 | hes | 4 378 7 | 123 20 1 | 523 359 25 |
56 | help | 2 609 8 | -820 515 29 | 349 94 7 |
57 | hllr | 11 300 8 | 2 0 0 | -295 300 24 |
58 | ill | 2 118 3 | 194 82 2 | 128 36 1 |
59 | im | 9 57 5 | 75 30 1 | 71 27 1 |
60 | ive | 4 232 5 | -54 7 0 | 308 226 10 |
61 | impr | 2 130 3 | -231 105 2 | 113 25 1 |
62 | indb | 3 13 49 | 30 0 0 | 280 13 6 |
63 | is | 35 352 8 | -163 323 17 | 50 30 2 |
64 | isis | 2 10 12 | 131 9 1 | -47 1 0 |
65 | its | 20 418 6 | 213 413 17 | -22 5 0 |
66 | job | 3 28 5 | -131 28 1 | 18 1 0 |
67 | jobs | 9 423 11 | -78 14 1 | -429 409 42 |
68 | just | 12 273 4 | -45 17 0 | 177 256 9 |
69 | keep | 2 77 4 | -224 75 2 | -38 2 0 |
70 | know | 19 610 7 | 76 40 2 | 285 569 39 |
71 | lght | 2 165 7 | 30 1 0 | 436 165 11 |
72 | let | 3 3 3 | 17 1 0 | 27 2 0 |
73 | like | 10 284 3 | 183 276 6 | 31 8 0 |
74 | look | 7 265 7 | 288 214 10 | 142 52 3 |
75 | lot | 6 346 4 | 187 127 4 | 246 220 9 |
76 | love | 4 29 5 | 121 29 1 | -3 0 0 |
77 | made | 3 3 3 | -36 3 0 | -12 0 0 |
78 | make | 9 370 5 | -290 365 13 | 34 5 0 |
79 | mean | 4 415 6 | 392 227 10 | 357 188 11 |
80 | mexc | 2 308 5 | 444 222 8 | -276 86 4 |
81 | mltr | 2 6 12 | 43 1 0 | -98 5 1 |
82 | mlln | 2 7 5 | 34 2 0 | 66 6 0 |
83 | mony | 4 300 4 | 323 279 8 | -89 21 1 |
84 | need | 4 249 6 | -335 182 8 | 203 67 4 |
85 | new | 6 123 12 | -225 73 6 | -186 50 5 |
86 | not | 18 316 4 | -139 226 6 | 88 90 3 |
87 | obam | 3 18 5 | 89 12 0 | -67 7 0 |
88 | ok | 4 597 8 | 644 488 28 | 304 109 8 |
89 | only | 3 207 5 | -262 125 4 | -212 82 4 |
90 | out | 12 90 4 | -26 6 0 | 101 84 3 |
91 | pay | 3 130 6 | -237 75 3 | 203 55 3 |
92 | pepl | 25 10 5 | 18 5 0 | 20 5 0 |
93 | prcn | 5 435 6 | 206 84 4 | -420 350 20 |
94 | ph | 2 136 10 | 353 74 5 | 325 62 6 |
95 | plac | 2 5 3 | -12 0 0 | -47 4 0 |
96 | plan | 2 184 7 | -360 113 6 | -285 71 5 |
97 | prsd | 6 539 10 | -483 358 26 | 343 181 17 |
98 | put | 3 34 2 | -91 30 1 | -30 3 0 |
99 | rlly | 6 427 6 | -128 43 2 | 381 384 20 |
100 | rmmb | 3 103 4 | 67 9 0 | -220 94 4 |
101 | rght | 11 294 4 | 201 286 9 | 35 9 0 |
102 | said | 12 567 10 | 275 236 17 | 326 331 32 |
103 | say | 8 269 3 | 141 144 3 | 132 125 4 |
104 | see | 6 233 4 | 159 99 3 | 185 134 5 |
105 | seen | 2 15 4 | 88 13 0 | -36 2 0 |
106 | she | 15 307 13 | 209 133 13 | -239 174 22 |
107 | shes | 4 273 5 | 358 237 9 | -140 36 2 |
108 | stat | 4 192 5 | -143 43 2 | -265 149 7 |
109 | stts | 4 47 3 | -21 2 0 | -105 45 1 |
110 | stop | 2 223 4 | 160 41 1 | -335 181 6 |
111 | take | 6 74 3 | 127 74 2 | 1 0 0 |
112 | talk | 2 112 4 | -73 8 0 | 271 105 4 |
113 | tax | 2 36 11 | -251 34 3 | -67 2 0 |
114 | tell | 5 56 3 | -13 1 0 | 112 55 2 |
115 | thank | 9 171 13 | -252 110 10 | -186 60 7 |
116 | thts | 10 84 4 | 45 15 0 | 98 69 2 |
117 | thrs | 2 31 3 | 66 8 0 | 110 23 1 |
118 | thes | 7 120 5 | 157 93 3 | -85 27 1 |
119 | thyr | 10 623 10 | 496 623 45 | -13 0 0 |
120 | thing | 3 358 4 | 293 202 5 | 257 156 5 |
121 | thngs | 4 128 3 | 130 66 1 | 125 62 1 |
122 | think | 9 440 6 | 99 37 2 | 325 402 23 |
123 | thos | 5 236 4 | -275 230 6 | 47 7 0 |
124 | time | 7 35 4 | -69 23 1 | -50 12 0 |
125 | tgth | 3 670 12 | -944 606 51 | 305 63 7 |
126 | too | 2 95 4 | -231 86 2 | 74 9 0 |
127 | trad | 3 253 8 | 275 86 5 | -384 167 13 |
128 | trmp | 18 16 17 | 74 16 2 | -9 0 0 |
129 | untd | 4 122 4 | -103 29 1 | -184 93 3 |
130 | up | 10 154 3 | -72 41 1 | 121 113 4 |
131 | very | 12 126 9 | 181 109 7 | -72 17 1 |
132 | vote | 5 147 10 | -330 136 10 | -98 12 1 |
133 | wall | 3 311 6 | 248 83 4 | -412 228 13 |
134 | want | 12 245 7 | -151 103 5 | 178 143 10 |
135 | wnts | 3 98 3 | 128 39 1 | -156 59 2 |
136 | was | 18 392 10 | 83 31 2 | 283 362 35 |
137 | way | 6 125 2 | 130 121 2 | 25 5 0 |
138 | well | 3 143 5 | 200 60 2 | -235 83 4 |
139 | were | 22 374 12 | 249 312 26 | -111 62 7 |
140 | well | 5 452 6 | -36 3 0 | 440 449 25 |
141 | were | 6 108 6 | -74 13 1 | 197 94 5 |
142 | whts | 4 331 3 | 339 329 8 | -26 2 0 |
143 | why | 3 81 4 | -111 28 1 | 151 53 2 |
144 | win | 4 222 8 | 297 124 7 | -264 98 8 |
145 | work | 5 553 13 | -727 504 47 | 226 49 6 |
146 | wrkn | 3 297 6 | -480 288 11 | 83 9 0 |
147 | wrld | 4 62 3 | -84 26 1 | -98 35 1 |
148 | yers | 6 98 3 | 91 41 1 | -108 57 2 |
149 | youre | 5 184 4 | 240 154 5 | 105 30 1 |
150 | your | 11 273 7 | -211 197 9 | -132 77 5 |Linguistic example
Basic plot
Linguistic example
Customized plot
Code
texts_df <- row_pcoord(d_ca)[,c(1, 2)] %>%
as_tibble(rownames = "text") %>%
mutate(Subcorpus = re_retrieve_first(fnames, "/clinton_trump/([^/]+)", requested_group = 1))
words_df <- col_pcoord(d_ca)[,c(1, 2)] %>%
as_tibble(rownames = "word")
dim_1 <- sprintf("Dimension 1 (%.2f %%)", summary(d_ca)$scree[1,3])
dim_2 <- sprintf("Dimension 2 (%.2f %%)", summary(d_ca)$scree[2,3])
ggplot(words_df, aes(x = V1, y = V2)) +
geom_text(aes(label = word), color = "gray60") +
geom_point(data = texts_df, aes(color = Subcorpus)) +
scale_color_manual(values = c("#0000CD","#DC143C")) +
geom_hline(yintercept = 0, color = "darkgray") +
geom_vline(xintercept = 0, color = "darkgray") +
theme_bw(base_size = 12) +
labs(x = dim_1, y = dim_2) +
coord_fixed()
Linguistic example
Next: Factor Analysis
Correspondence analysis Methods of Corpus Linguistics (class 7) Mariana Montes