statistics

temporal distribution

We plot some temporal distributions:

echo "year articles found true prec"
for year in $(seq 1987 2007); do
	echo $year \
	 $(grep ^$year articles.tsv | cut -d' ' -f2) \
	 $(../org.py -f year README.org | grep ${year} | wc -l) \
		 $(../org.py -f year,classification,status README.org | grep ${year} | awk -F$'\t' '{if ($3 == "D" || $2 == "True") print;}' | wc -l)
done

year	articles	found	true	prec	ppm
1987	106104	207	103	49.8	0.97
1988	104541	223	99	44.4	0.95
1989	102818	227	109	48.0	1.06
1990	98812	232	111	47.8	1.12
1991	85135	217	107	49.3	1.26
1992	82685	230	115	50.0	1.39
1993	79200	239	124	51.9	1.57
1994	74925	252	129	51.2	1.72
1995	85392	249	134	53.8	1.57
1996	79077	306	155	50.7	1.96
1997	85396	278	143	51.4	1.67
1998	89163	338	191	56.5	2.14
1999	91074	320	150	46.9	1.65
2000	94258	362	188	51.9	1.99
2001	96282	319	165	51.7	1.71
2002	97258	389	191	49.1	1.96
2003	94235	357	186	52.1	1.97
2004	91362	339	163	48.1	1.78
2005	90004	396	179	45.2	1.99
2006	87052	411	187	45.5	2.15
2007	39953	180	85	47.2	2.13
sum	1854726	6071	3014	49.6	34.71
mean	88320	289	144	49.8	1.63

: The temporal distribution of the number of found and true candidates.

reset
set datafile separator "\t"

set xlabel "year"
set ylabel "frequency"
set grid linetype 1 linecolor 0
set yrange [0:*]
set y2range [0:100]
set y2label 'precision'
set y2tics
set key top left
set style fill solid 1

set term svg enhanced size 800,600 dynamic fname "Palatino Linotype, Book Antiqua, Palatino, FreeSerif, serif" fsize 16
#set out "nyt_vossantos_over_time.svg"
plot data using 1:3 with linespoints pt 7 lc "red" title 'candidates',\
	 data using 1:4 with linespoints pt 7 lc "green"  title 'Vossantos',\
	 data using 1:5 with lines lc "blue" title 'precision' axes x1y2

# data using 1:2 with linespoints pt 7 axes x1y2 title 'cand',\
#     data using 1:3 with linespoints pt 7 axes x1y2 title 'wd',\

set term png enhanced size 800,600 font "Arial,16" lw  2
set out "nyt_vossantos_over_time.png"
replot

set key bottom left
set term pdf enhanced fontscale .7 lw 2
set out "nyt_vossantos_over_time.pdf"
replot


# ---- relative values

set key top left
set term svg enhanced size 800,600 dynamic fname "Palatino Linotype, Book Antiqua, Palatino, FreeSerif, serif" fsize 16
set out "nyt_vossantos_over_time_rel.svg"
set ylabel "frequency (per mille)"
set format y "%2.1f"

plot data using 1:($3/$2*1000) with linespoints pt 7 lc "red" title 'candidates',\
	 data using 1:($4/$2*1000) with linespoints pt 7 lc "green"  title 'Vossantos',\
	 data using 1:5 with lines lc "blue" title 'precision' axes x1y2


set term png enhanced size 800,600 font "Arial,16" lw  2
set out "nyt_vossantos_over_time_rel.png"
replot

set term pdf enhanced lw  2
set out "nyt_vossantos_over_time_rel.pdf"
replot

Absolute frequency:

Relative frequency:

sources

most frequent

The most frequent sources are:

../org.py -T -f sourceUrl README.org | sort | uniq -c | sort -nr | head -n40

count	source
72	Michael Jordan
62	Rodney Dangerfield
40	Johnny Appleseed
36	Elvis Presley
36	Babe Ruth
25	Michelangelo
25	Donald Trump
23	Pablo Picasso
23	Bill Gates
23	Madonna
21	Jackie Robinson
20	P. T. Barnum
20	Tiger Woods
19	Martha Stewart
17	William Shakespeare
17	Wolfgang Amadeus Mozart
17	Cinderella
16	Henry Ford
16	John Wayne
15	Napoleon
14	Leonardo da Vinci
14	Greta Garbo
14	Rosa Parks
14	Adolf Hitler
14	Mother Teresa
14	Ralph Nader
13	Cal Ripken
12	Willie Horton
12	Leo Tolstoy
12	Rembrandt
12	Oprah Winfrey
12	Susan Lucci
11	Walt Disney
11	Mike Tyson
10	Albert Einstein
10	Thomas Edison
10	Paul Revere
10	Julia Child
10	Cassandra
9	James Dean

temporal distribution

for year in $(seq 1987 2007); do
  echo -n $year
  for s in "Michael_Jordan" "Rodney_Dangerfield" "Johnny_Appleseed"; do
	s=$(echo $s| sed "s/_/ /g")
	c=$(../org.py -T -f year,sourceLabel README.org | grep ^$year | awk -F'\t' '{print $2}' | grep "^$s$" | wc -l)
	echo -n "\t$c"
  done
  echo
done

year	Michael Jordan	Rodney Dangerfield	Johnny Appleseed
1987	0	0	2
1988	0	0	1
1989	1	1	1
1990	3	2	1
1991	4	1	1
1992	2	4	1
1993	3	4	2
1994	3	0	0
1995	0	1	3
1996	4	8	3
1997	1	3	1
1998	6	7	2
1999	11	2	3
2000	11	6	1
2001	7	5	1
2002	5	2	3
2003	2	1	3
2004	0	1	3
2005	2	8	4
2006	4	5	3
2007	3	1	1

reset
set datafile separator "\t"

set xlabel "year"
set ylabel "frequency"
set grid linetype 1 linecolor 0
set yrange [0:*]
set key top left
set style fill solid 1

set term svg enhanced size 800,600 dynamic font "Palatino Linotype, 16"
#set out "nyt_sources_over_time.svg"
plot data using 1:2 with linespoints pt 7 lw 2 title 'Michael Jordan',\
	 data using 1:3 with linespoints pt 7 title 'Rodney Dangerfield',\
	 data using 1:4 with linespoints pt 7 title 'Johnny Appleseed'

set term png enhanced size 800,600 font "Arial,16" lw  2
set out "nyt_sources_over_time.png"
replot

categories

online

Extract the categories for the articles:

export PYTHONIOENCODING=utf-8
for year in $(seq 1987 2007); do
	../nyt.py --category ../nyt_corpus_${year}.tar.gz \
		| sed -e "s/^nyt_corpus_//" -e "s/\.har\//\//" -e "s/\.xml\t/\t/" \
		| sort >> nyt_categories.tsv
done

Compute frequency distribution over all articles:

cut -d$'\t' -f2 nyt_categories.tsv | sort -S1G | uniq -c \
   | sed -e "s/^ *//" -e "s/ /\t/" | awk -F'\t' '{print $2"\t"$1}' \
										  > nyt_categories_distrib.tsv

Check the number of and the top categories:

echo articles $(wc -l < nyt_categories.tsv)
echo categories $(wc -l < nyt_categories_distrib.tsv)
echo ""
sort -nrk2 nyt_categories_distrib.tsv | head

articles	1854726
categories	1580
Business	291982
Sports	160888
Opinion	134428
U.S.	89389
Arts	88460
World	79786
Style	65071
Obituaries	19430
Magazine	11464
Travel	10440

Collect the categories of the articles

echo "vossantos" $(../org.py -T README.org | wc -l) articles $(wc -l < nyt_categories.tsv)
../org.py -T -f fId README.org | join nyt_categories.tsv - | sed "s/ /\t/" | awk -F'\t' '{print $2}' \
	| sort | uniq -c \
	| sed -e "s/^ *//" -e "s/ /\t/" | awk -F'\t' '{print $2"\t"$1}' \
	| join -t$'\t' -o1.2,1.1,2.2 - nyt_categories_distrib.tsv \
	| sort -nr | head -n20

vossantos	3014	category	articles	1854726
364	12.1%	Arts	88460	4.8%
362	12.0%	Sports	160888	8.7%
327	10.8%	New York and Region	221897	12.0%
287	9.5%	Arts; Books	35475	1.9%
186	6.2%	Movies; Arts	27759	1.5%
125	4.1%	Business	291982	15.7%
122	4.0%	Opinion	134428	7.2%
110	3.6%	U.S.	89389	4.8%
104	3.5%	Magazine	11464	0.6%
76	2.5%	Arts; Theater	13283	0.7%
70	2.3%	Style	65071	3.5%
52	1.7%	World	79786	4.3%
49	1.6%	Home and Garden; Style	13978	0.8%
37	1.2%		42157	2.3%
36	1.2%	Travel	10440	0.6%
35	1.2%	Technology; Business	23283	1.3%
30	1.0%	Week in Review	17107	0.9%
29	1.0%	Home and Garden	5546	0.3%
18	0.6%	Style; Magazine	1519	0.1%
18	0.6%	Front Page; U.S.	11425	0.6%

desks

Extract the desks for the articles:

export PYTHONIOENCODING=utf-8
for year in $(seq 1987 2007); do
	../nyt.py --desk ../nyt_corpus_${year}.tar.gz \
		| sed -e "s/^nyt_corpus_//" -e "s/\.har\//\//" -e "s/\.xml\t/\t/" \
		| sort >> nyt_desks.tsv
done

Compute frequency distribution over all articles:

cut -d$'\t' -f2 nyt_desks.tsv | sort -S1G | uniq -c \
   | sed -e "s/^ *//" -e "s/ /\t/" | awk -F'\t' '{print $2"\t"$1}' \
										  > nyt_desks_distrib.tsv

Check the number of and the top categories:

echo articles $(wc -l < nyt_desks.tsv)
echo categories $(wc -l < nyt_desks_distrib.tsv)
echo ""
sort -t$'\t' -nrk2 nyt_desks_distrib.tsv | head

articles	1854727
categories	398
Metropolitan Desk	237896
Financial Desk	206958
Sports Desk	174823
National Desk	143489
Editorial Desk	131762
Foreign Desk	129732
Classified	129660
Business/Financial Desk	112951
Society Desk	44032
Cultural Desk	40342

Collect the desks of the articles

echo "vossantos" $(../org.py -T README.org | wc -l) articles $(wc -l < nyt_desks.tsv)
../org.py -T -f fid README.org | join nyt_desks.tsv - | sed "s/ /\t/" | awk -F'\t' '{print $2}' \
	| sort | uniq -c \
	| sed -e "s/^ *//" -e "s/ /\t/" | awk -F'\t' '{print $2"\t"$1}' \
	| join -t$'\t' -o1.2,1.1,2.2 - nyt_desks_distrib.tsv \
	| sort -nr | head -n20

vossantos	3014	desk	articles	1854726
381	12.6%	Sports Desk	174823	9.4%
222	7.4%	Metropolitan Desk	237896	12.8%
220	7.3%	Book Review Desk	32737	1.8%
180	6.0%	National Desk	143489	7.7%
171	5.7%	The Arts/Cultural Desk	38136	2.1%
169	5.6%	Arts and Leisure Desk	27765	1.5%
135	4.5%	Magazine Desk	25433	1.4%
125	4.1%	Editorial Desk	131762	7.1%
117	3.9%	Cultural Desk	40342	2.2%
99	3.3%	Movies, Performing Arts/Weekend Desk	13929	0.8%
96	3.2%	Business/Financial Desk	112951	6.1%
90	3.0%	Foreign Desk	129732	7.0%
78	2.6%	Weekend Desk	18814	1.0%
74	2.5%	Leisure/Weekend Desk	10766	0.6%
72	2.4%	Long Island Weekly Desk	20453	1.1%
69	2.3%	Style Desk	21569	1.2%
57	1.9%	Financial Desk	206958	11.2%
44	1.5%	Arts & Leisure Desk	6742	0.4%
42	1.4%	The City Weekly Desk	22863	1.2%
41	1.4%	Connecticut Weekly Desk	17034	0.9%

Note: there are many errors in the specification of the desks … so this table should be digested with care.

authors

Extract the authors for the articles:

export PYTHONIOENCODING=utf-8
for year in $(seq 1987 2007); do
	../nyt.py --author ../nyt_corpus_${year}.tar.gz \
		| sed -e "s/^nyt_corpus_//" -e "s/\.har\//\//" -e "s/\.xml\t/\t/" \
		| sort >> nyt_authors.tsv
done

Compute frequency distribution over all articles:

cut -d$'\t' -f2 nyt_authors.tsv | LC_ALL=C sort -S1G | uniq -c \
   | sed -e "s/^ *//" -e "s/ /\t/" | awk -F'\t' '{print $2"\t"$1}' \
										  > nyt_authors_distrib.tsv

Check the number of and the top authors:

echo articles $(wc -l < nyt_authors.tsv)
echo categories $(wc -l < nyt_authors_distrib.tsv)
echo ""
sort -t$'\t' -nrk2 nyt_authors_distrib.tsv | head

articles	1854726
categories	30691
	961052
Elliott, Stuart	6296
Holden, Stephen	5098
Chass, Murray	4544
Pareles, Jon	4090
Brozan, Nadine	3741
Fabricant, Florence	3659
Kozinn, Allan	3654
Curry, Jack	3654
Truscott, Alan	3646

requires cleansing!

Collect the authors of the articles

echo "vossantos" $(../org.py -T README.org | wc -l) articles $(wc -l < nyt_authors.tsv)
../org.py -T -f fid README.org | join nyt_authors.tsv - | sed "s/ /\t/" | awk -F'\t' '{print $2}' \
	| LC_ALL=C sort | uniq -c \
	| sed -e "s/^ *//" -e "s/ /\t/" | awk -F'\t' '{print $2"\t"$1}' \
	| LC_ALL=C join -t$'\t' -o1.2,1.1,2.2 - nyt_authors_distrib.tsv \
	| sort -nr | head -n20

vossantos	3014	author	articles	1854726
470	15.6%		961052	51.8%
34	1.1%	Maslin, Janet	2874	0.2%
32	1.1%	Holden, Stephen	5098	0.3%
30	1.0%	Vecsey, George	2739	0.1%
24	0.8%	Sandomir, Richard	3140	0.2%
24	0.8%	Dowd, Maureen	1647	0.1%
23	0.8%	Ketcham, Diane	717	0.0%
20	0.7%	Kisselgoff, Anna	2661	0.1%
20	0.7%	Brown, Patricia Leigh	568	0.0%
19	0.6%	Kimmelman, Michael	1515	0.1%
19	0.6%	Berkow, Ira	1704	0.1%
18	0.6%	Barron, James	2188	0.1%
17	0.6%	Stanley, Alessandra	1437	0.1%
17	0.6%	Pareles, Jon	4090	0.2%
17	0.6%	Lipsyte, Robert	817	0.0%
17	0.6%	Araton, Harvey	1940	0.1%
16	0.5%	Smith, Roberta	2497	0.1%
16	0.5%	Martin, Douglas	1814	0.1%
16	0.5%	Chass, Murray	4544	0.2%
15	0.5%	Grimes, William	1368	0.1%

Vossantos of the top author

# extract list of articles
for article in $(../org.py -T -f fid README.org | join nyt_authors.tsv - | grep "Maslin, Janet" | cut -d' ' -f1 ); do
  grep "$article" README.org
done

1. Bob Hope (1993/04/23/0604282) is loaded with rap-related cameos that work only if you recognize the players (Fab 5 Freddy, Kid Capri, Naughty by Nature and the Bob Hope of rap cinema, Ice-T), and have little intrinsic humor of their own.
2. Sandy Dennis (1993/09/03/0632371) (Ms. Lewis, who has many similar mannerisms, may be fast becoming the Sandy Dennis of her generation.)
3. Dorian Gray (1993/12/10/0654992) Also on hand is Aerosmith, the Dorian Gray of rock bands, to serve the same purpose Alice Cooper did in the first film.
4. Adolf Hitler (1994/02/04/0666537) The terrors of the code, as overseen by Joseph Breen (who was nicknamed “the Hitler of Hollywood” in some quarters), went beyond the letter of the document and brought about a more generalized moral purge.
5. Cinderella (1994/09/11/0711230) Kevin Smith, the Cinderella of this year's Sundance festival, shot this black-and-white movie in the New Jersey store where he himself worked.
6. Hulk Hogan (1994/10/25/0720551) Libby's cousin Andrew, an art director who's “so incredibly creative that, as my mother says, no one's holding their breath for grandchildren,” opines that “David Mamet is the Hulk Hogan of the American theater and that his word processor should be tested for steroids.”
7. Andrew Dice Clay (1995/09/22/0790066) Mr. Ezsterhas, the Andrew Dice Clay of screenwriting, bludgeons the audience with such tirelessly crude thoughts that when a group of chimps get loose in the showgirls' dressing room and all they do is defecate, the film enjoys a rare moment of good taste.
8. Thomas Jefferson (1996/01/24/0825044) Last year's overnight sensation, Edward Burns of “The Brothers McMullen,” came out of nowhere and now has Jennifer Aniston acting in his new film and Robert Redford, the Thomas Jefferson of Sundance, helping as a creative consultant.
9. Elliott Gould (1996/03/08/0835139) All coy grins and daffy mugging, Mr. Stiller plays the role as if aspiring to become the Elliott Gould of his generation.
10. Charlie Parker (1996/08/09/0870295) But for all its admiration, ‘'Basquiat’' winds up no closer to that assessment than to the critic Robert Hughes's more jaundiced one: ‘'Far from being the Charlie Parker of SoHo (as his promoters claimed), he became its Jessica Savitch.’'
11. Aesop (1996/08/09/0870300) Eric Rohmer's ‘'Rendezvous in Paris’' is an oasis of contemplative intelligence in the summer movie season, presenting three graceful and elegant parables with the moral agility that distinguishes Mr. Rohmer as the Aesop of amour.
12. Diana Vreeland (1997/06/06/0934955) The complex aural and visual style of ‘'The Pillow Book’' involves rectangular insets that flash back to Sei Shonagon (a kind of Windows 995) and illustrate the imperious little lists that made her sound like the Diana Vreeland of 10th-century tastes.
13. Peter Pan (1997/08/08/0949060) Mr. Gibson, delivering one of the hearty, dynamic star turns that have made him the Peter Pan of the blockbuster set, makes Jerry much more boyishly likable than he deserves to be.
14. Thomas Edison (1997/09/19/0958685) Danny DeVito embodies this as a gleeful Sid Hudgens (a character whom Mr. Hanson has called '‘the Thomas Edison of tabloid journalism’’), who is the unscrupulous editor of a publication called Hush-Hush and winds up linked to many of the other characters’ nastiest transgressions.
15. John Wayne (1997/09/26/0960422) Mr. Hopkins, whose creative collaboration with Bart goes back to ‘'Legends of the Fall,’' has called him '‘the John Wayne of bears.’'
16. Annie Oakley (1997/12/24/0982708) Running nearly as long as ‘'Pulp Fiction’' even though its ambitions are more familiar and small, ‘'Jackie Brown’' has the makings of another, chattier ‘'Get Shorty’' with an added homage to Pam Grier, the Annie Oakley of 1970's blaxploitation.
17. Robin Hood (1998/04/10/1008616) Do not threaten to call the police or have him thrown out,'' went a memorandum issued by another company, when the Robin Hood of corporate America went on the road to promote his book abou downsizing.
18. Buster Keaton (1998/09/18/1047276) Fortunately, being the Buster Keaton of martial arts, he makes a doleful expression and comedic physical grace take the place of small talk.
19. Michelangelo (1998/09/25/1049076) She goes to a plastic surgeon (Michael Lerner) who's been dubbed '‘the Michelangelo of Manhattan’' by Newsweek.
20. Brian Wilson (1998/12/31/1073562) The enrapturing beauty and peculiar naivete of ‘'The Thin Red Line’' heightened the impression of Terrence Malick as the Brian Wilson of the film world.
21. Dante Alighieri (1999/10/22/1147181) Though his latest film explores one more urban inferno and colorfully reaffirms Mr. Scorsese's role as the Dante of the Cinema, creating its air of nocturnal torment took some doing.
22. Albert Einstein (2000/12/07/1253134) In this much coarser and more violent, action-heavy story, Mr. Deaver presents the villainous Dr. Aaron Matthews, whom a newspaper once called '‘the Einstein of therapists’' in the days before Hannibal Lecter became his main career influence.
23. Émile Zola (2001/03/09/1276449) George P. Pelecanos arrives with the best possible recommendations from other crime writers (e.g., Elmore Leonard likes him), and with jacket copy praising him as '‘the Zola of Washington, D.C.’' But what he really displays here, in great abundance and to entertaining effect, is a Tarantino touch.
24. Leonard Cohen (2002/08/22/1417676) The wry, sexy melancholy of his observations would be seductive enough in its own right – he is the Leonard Cohen of the spy genre – even without the sharp political acuity that accompanies it.
25. Jane Austen (2002/10/07/1429887) Ms. Pearson does so well in capturing the funny, calculating aspects of her English heroine's life that The Guardian has called her '‘a Jane Austen among working mothers.’'
26. Kato Kaelin (2003/04/07/1478881) Then he has settled in – as ‘'a permanent house guest, the Kato Kaelin of the wine country,’' in the case of Alan Deutschman – and tried to figure out what it all means.
27. Hulk Hogan (2003/04/14/1480850) Meanwhile, at 5 feet 10 tall and 115 pounds, Andy is the Hulk Hogan of this food-phobic crowd.
28. Nora Roberts (2003/04/17/1481531) For those who write like clockwork (i.e., Stuart Woods, the Nora Roberts of mystery best-sellerdom), a new book every few months is no surprise.
29. Henny Youngman (2004/03/05/1563840) Together Mr. Yetnikoff and Mr. Ritz devise a kind of sitcom snappiness that turns Mr. Yetnikoff into the Henny Youngman of CBS.
30. Frank Stallone (2004/09/20/1612886) He can read the biblical story of Aaron and imagine '‘the Frank Stallone of ancient Judaism.’'
31. Marlon Brando (2005/11/08/1715899) He named his daughter Tuesday, after the actress Tuesday Weld, whom Sam Shepard once called '‘the Marlon Brando of women.’'
32. Jesse James (2005/12/09/1723424) How else to explain ‘'Comma Sense,’' which has a blurb from Ms. Truss and claims that the apostrophe is the Jesse James of punctuation marks?
33. Elton John (2006/12/11/1811150) Though Foujita had a fashion sense that made him look like the Elton John of Montparnasse (he favored earrings, bangs and show-stopping homemade costumes), and though he is seen here hand in hand with a male Japanese friend during their shared tunic-wearing phase, he is viewed by Ms. Birnbaum strictly as a lady-killer.
34. Ernest Hemingway (2007/04/30/1844006) Mr. Browne also points out that when he introduced Mr. Zevon to an audience as '‘the Ernest Hemingway of the twelve-string guitar,’' Mr. Zevon said he was more like Charles Bronson.

modifiers

../org.py -T -f modifier,aId README.org \
	| awk -F$'\t' '$1 != "" {print $1;}' \
	| sort | uniq -c | sort -nr | head -n30

count	modifier
56	his day
34	his time
29	Japan
17	China
16	tennis
16	his generation
16	baseball
14	her time
13	our time
13	her day
12	the Zulus
11	the 90's
11	the 1990's
11	politics
11	hockey
10	the art world
10	Brazil
10	basketball
10	ballet
9	jazz
9	fashion
8	today
8	Iran
8	his era
8	hip-hop
8	golf
8	football
8	dance
7	the 19th century
7	Mexico

today

“today”

Who are the sources for the modifier “today”?

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "today" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
1	Shoeless Joe Jackson
1	Buck Rogers
1	Bill McGowan
1	William F. Buckley Jr.
1	Ralph Fiennes
1	Julie London
1	Jimmy Osmond
1	Harry Cohn

“his day”, “his time”, or “his generation”

Who are the sources for the modifiers “his day”, “his time”, and “his generation”?

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 ~ "his (day|time|generation)" {print $2;}' \
	| sort | uniq -c | sort -nr | head

count	source
3	Donald Trump
2	Mike Tyson
2	Pablo Picasso
2	Billy Martin
2	Dan Quayle
2	Arnold Schwarzenegger
2	Martha Stewart
2	L. Ron Hubbard
2	Tiger Woods

“her day”, “her time”, or “her generation”

Who are the sources for the modifiers “her day”, “her time”, and “her generation”?

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 ~ "her (day|time|generation)" {print $2;}' \
	| sort | uniq -c | sort -nr | head

count	source
4	Madonna
2	Laurie Anderson
1	Hilary Swank
1	Pamela Anderson
1	Hillary Clinton
1	Lotte Lehmann
1	Oprah Winfrey
1	Marilyn Monroe
1	Coco Chanel
1	Judith Krantz

country

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 ~ "(Japan|China|Brazil|Iran|Israel|Mexico|India|South Africa|Spain|South Korea|Russia|Poland|Pakistan)" {print $1;}' \
	| sort | uniq -c | sort -nr | head

count	country
29	Japan
17	China
10	Brazil
8	Iran
7	Mexico
7	Israel
7	India
4	South Africa
4	Poland
3	Spain

What are the sources for the modifier … ?

“Japan”

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "Japan" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
5	Walt Disney
4	Bill Gates
2	Nolan Ryan
2	Frank Sinatra
1	Richard Perle
1	Thomas Edison
1	Cal Ripken
1	Walter Johnson
1	Andy Warhol
1	Pablo Picasso
1	William Wyler
1	Stephen King
1	Brad Pitt
1	Richard Avedon
1	P. D. James
1	Rem Koolhaas
1	Steve Jobs
1	Ralph Nader
1	Madonna
1	Jack Kerouac

“China”

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "China" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
4	Barbara Walters
2	Jack Welch
2	Larry King
1	Louis XIV of France
1	Oskar Schindler
1	Napoleon
1	Keith Haring
1	Mikhail Gorbachev
1	Donald Trump
1	Ted Turner
1	Madonna
1	The Scarlet Pimpernel

“Brazil”

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "Brazil" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
1	Giuseppe Verdi
1	Jil Sander
1	Walter Reed
1	Lech Wałęsa
1	Jim Morrison
1	Bob Dylan
1	Elvis Presley
1	Scott Joplin
1	Larry Bird
1	Pablo Escobar

sports

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 ~ "(baseball|hockey|basketball|tennis|golf|football|racing|soccer|sailing)" {print $1;}' \
	| sort | uniq -c | sort -nr

count	sports
16	tennis
16	baseball
11	hockey
10	basketball
8	golf
8	football
6	soccer
6	racing
3	women’s basketball
3	sailing
3	auto racing
2	pro football
2	New York baseball
1	Yale football fame
1	women’s hockey
1	women’s college soccer
1	this year’s national collegiate basketball tournament
1	the tennis tour
1	the tennis field
1	the soccer set
1	the racing world
1	the Olympic hockey tournament
1	stock-car racing
1	Rotisserie baseball
1	pro football owners
1	professional basketball coaches
1	professional basketball
1	motocross racing in the 1980’s
1	micro golfers
1	major league baseball
1	Laser sailing
1	Japanese baseball
1	Iraqi soccer
1	horse racing
1	hockey in the former Soviet Union
1	hockey commentary
1	high school baseball in New York
1	harness racing
1	golf criticism
1	football teams
1	football owners
1	football announcers
1	European hockey
1	country-club golf
1	college football underclassmen
1	college football these days
1	college football
1	college basketball
1	Chinese baseball
1	Brazilian basketball for the past 20 years
1	BMX racing
1	biddy basketball
1	basketball announcers
1	basketball analysts
1	basketball analysis
1	baseball’s new era
1	baseball managers
1	baseball executives
1	baseball collections
1	baseball cards

Who are the sources for the modifier … ?

baseball

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "baseball" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
2	P. T. Barnum
2	Larry Bird
1	Clifford Irving
1	Mike Tyson
1	Thomas Dooley
1	Marco Polo
1	Pablo Picasso
1	Horatio Alger
1	Rodney Dangerfield
1	Michael Jordan
1	Alan Alda
1	Brandon Tartikoff
1	Howard Hughes
1	Thomas Jefferson

tennis

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "tennis" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
2	George Foreman
1	Tim McCarver
1	Pete Rose
1	Nolan Ryan
1	Crash Davis
1	Spike Lee
1	John Madden
1	Michael Jordan
1	John Wayne
1	George Hamilton
1	Michael Dukakis
1	Jackie Robinson
1	Babe Ruth
1	Dennis Rodman
1	Madonna

basketball

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "basketball" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
2	Babe Ruth
1	Joseph Stalin
1	Martin Luther King, Jr.
1	Pol Pot
1	Johnny Appleseed
1	Adolf Hitler
1	Bugsy Siegel
1	Elvis Presley
1	Chuck Yeager

football

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "football" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
1	Ann Calvello
1	Michael Jordan
1	Bobby Fischer
1	Patrick Henry
1	Susan Lucci
1	Jackie Robinson
1	Babe Ruth
1	Rich Little

racing

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "racing" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
2	Rodney Dangerfield
1	John Madden
1	Bobo Holloman
1	Lou Gehrig
1	Wayne Gretzky

golf

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 == "golf" {print $2;}' \
	| sort | uniq -c | sort -nr

count	source
2	Michael Jordan
2	Jackie Robinson
1	J. D. Salinger
1	James Brown
1	Marlon Brando
1	Babe Ruth

culture

../org.py -T -f modifier,sourceUrl README.org \
	| awk -F$'\t' '$1 ~ "(dance|hip-hop|jazz|fashion|weaving|ballet|the art world|wine|salsa|juggling|tango)" {print $1;}' \
	| sort | uniq -c | sort -nr | head -n13

count	modifier
10	the art world
10	ballet
9	jazz
9	fashion
8	hip-hop
8	dance
4	wine
4	salsa
2	the hip-hop world
2	the fashion world
2	the fashion industry
2	the dance world
2	juggling

Michael Jordan

../org.py -T -f sourceLabel,modifier README.org \
	| awk -F$'\t' '{if ($1 == "Michael Jordan") print $2}' \
	| sort -u

the Michael Jordan of

• …
• 12th men
• actresses
• Afghanistan
• Australia
• baseball
• BMX racing
• boxing
• Brazilian basketball for the past 20 years
• bull riding
• college coaches
• computer games
• cricket
• cyberspace
• dance
• diving
• dressage horses
• fast food
• figure skating
• foosball
• football
• game shows
• geopolitics
• golf
• Harlem
• her time
• his day
• his sport
• his team
• his time
• hockey
• horse racing
• hunting and fishing
• Indiana
• integrating insurance and health care
• julienne
• jumpers
• language
• Laser sailing
• late-night TV
• management in Digital
• Mexico
• motocross racing in the 1980's
• orange juice
• real-life bulls
• recording
• Sauternes
• snowboarding
• soccer
• television puppets
• tennis
• the Buffalo team
• the dirt set
• the Eagles
• the game
• the Hudson
• the National Football League
• the South Korean penal system
• the sport
• the White Sox
• this sport
• women's ball
• women's basketball