forked from aimacode/aima-julia
-
Notifications
You must be signed in to change notification settings - Fork 0
/
nlp.jl
788 lines (673 loc) · 27.7 KB
/
nlp.jl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
export Rules, Lexicon, Grammar,
rewrites_for, is_category, cnf_rules, rewrite, generate_random_sentence,
ProbabilityRules, ProbabilityLexicon, ProbabilityGrammar,
Chart, parse_sentence, cyk_parse,
Page, load_page_html, determine_inlinks, find_outlinks, init_pages, only_wikipedia_urls,
expand_pages, relevant_pages, normalize_pages,
ConvergenceDetector, detect_convergence, get_inlinks, get_outlinks, HITS;
"""
Rules(rules_array::Array{T}) where {T <: Pair}
Return a Dict of mappings for symbols (lexical categories) to alternative sequences.
"""
function Rules(rules_array::Array{T, 1}) where {T <: Pair}
local rules::Dict = Dict();
for (lhs, rhs) in rules_array
rules[lhs] = collect(map(String, split(strip(ss))) for ss in map(String, split(rhs, ['|'])));
end
return rules;
end
"""
Lexicon(rules_array::Array{T}) where {T <: Pair}
Return a Dict of mappings for symbols (lexical categories) to alternative words.
The lexicon is the list of allowable words.
"""
function Lexicon(rules_array::Array{T, 1}) where {T <: Pair}
local rules::Dict = Dict();
for (lhs, rhs) in rules_array
rules[lhs] = collect(strip(ss) for ss in map(String, split(rhs, "|")));
end
return rules;
end
#=
Grammar consists of a set of rules and a lexicon.
=#
struct Grammar
name::String
rules::Dict
lexicon::Dict
categories::Dict
function Grammar(name::String, rules::Dict, lexicon::Dict)
local ng::Grammar = new(name, rules, lexicon, Dict());
for (category, words) in ng.lexicon
for word in words
ng.categories[word] = push!(get!(ng.categories, word, Array{String, 1}()), category);
end
end
return ng;
end
end
"""
rewrites_for(g::Grammar, cat::String)
Return an Array of possible rhs's that category 'cat' can be rewritten as.
"""
function rewrites_for(g::Grammar, cat::String)
return get(g.rules, cat, Array{String, 1}());
end
"""
is_category(g::Grammar, word::String, cat::String)
Return whether the given word 'word' is of category 'cat'.
"""
function is_category(g::Grammar, word::String, cat::String)
return (cat in g.categories[word]);
end
"""
cnf_rules(g::Grammar)
Return the rules of grammar 'g' as an Array of Tuples (X, Y, Z) such that X -> Y Z.
"""
function cnf_rules(g::Grammar)
local cnf::AbstractVector = [];
for (x, rules) in g.rules
for (y, z) in rules
push!(cnf, (x, y, z));
end
end
return cnf;
end
"""
rewrite(g::Grammar, tokens::AbstractVector, into::AbstractVector)
Return the resulting array of words by replacing each token in 'tokens' with a random word.
"""
function rewrite(g::Grammar, tokens::AbstractVector, into::AbstractVector)
for token in tokens
if (token in keys(g.rules))
rewrite(g, rand(RandomDeviceInstance, g.rules[token]), into);
elseif (token in keys(g.lexicon))
push!(into, rand(RandomDeviceInstance, g.lexicon[token]));
else
push!(into, token);
end
end
return into;
end
"""
generate_random_sentence(g::Grammar, categories::String)
generate_random_sentence(g::Grammar)
Return a randomly generated sentence as a String by using the given categories 'categories'.
"""
function generate_random_sentence(g::Grammar, categories::String)
return join(rewrite(g, split(categories), []), " ");
end
function generate_random_sentence(g::Grammar)
return generate_random_sentence(g, "S");
end
"""
ProbabilityRules(rules_array::Array{T}) where {T <: Pair}
Return a Dict of mappings for symbols (lexical categories) to alternative sequences with probabilities.
"""
function ProbabilityRules(rules_array::Array{T, 1}) where {T <: Pair}
local rules::Dict = Dict();
for (lhs, rhs) in rules_array
rules[lhs] = [];
local rhs_split::AbstractVector = collect(map(String, split(strip(ss))) for ss in map(String, split(rhs, ['|'])));
for rule in rhs_split
local rule_probability::Float64 = parse(Float64, rule[end][2:(end - 1)]);
local rule_tuple::Tuple = (rule[1:(end - 1)], rule_probability);
push!(rules[lhs], rule_tuple);
end
end
return rules;
end
"""
ProbabilityLexicon(rules_array::Array{T}) where {T <: Pair}
Return a Dict of mappings for symbols (lexical categories) to alternative words with probabilities.
The lexicon is the list of allowable words.
"""
function ProbabilityLexicon(rules_array::Array{T, 1}) where {T <: Pair}
local rules::Dict = Dict();
for (lhs, rhs) in rules_array
rules[lhs] = [];
local rhs_split::AbstractVector = collect(map(String, split(strip(word))) for word in map(String, split(rhs, ['|'])));
for rule in rhs_split
local word_probability::Float64 = parse(Float64, rule[end][2:(end - 1)]);
local rule_word::String = rule[1];
local rule_tuple::Tuple = (rule_word, word_probability);
push!(rules[lhs], rule_tuple);
end
end
return rules;
end
#=
ProbabilityGrammar consists of a set of rules and a lexicon.
=#
struct ProbabilityGrammar
name::String
rules::Dict
lexicon::Dict
categories::Dict
function ProbabilityGrammar(name::String, rules::Dict, lexicon::Dict)
local npg::ProbabilityGrammar = new(name, rules, lexicon, Dict());
for (category, words) in npg.lexicon
for (word, p) in words
npg.categories[word] = push!(get!(npg.categories, word, []), (category, p));
end
end
return npg;
end
end
"""
rewrites_for(pg::ProbabilityGrammar, cat::String)
Return an Array of possible rhs's that category 'cat' can be rewritten as.
"""
function rewrites_for(pg::ProbabilityGrammar, cat::String)
return get(pg.rules, cat, []);
end
"""
is_category(pg::ProbabilityGrammar, word::String, cat::String)
Return whether the given word 'word' is of category 'cat'.
"""
function is_category(pg::ProbabilityGrammar, word::String, cat::String)
return (cat in map(first, pg.categories[word]));
end
"""
cnf_rules(pg::ProbabilityGrammar)
Return the rules of grammar 'pg' as an Array of Tuples (X, Y, Z, p) such that X -> Y Z [p].
"""
function cnf_rules(pg::ProbabilityGrammar)
local cnf::AbstractVector = [];
for (x, rules) in pg.rules
for ((y, z), p) in rules
push!(cnf, (x, y, z, p));
end
end
return cnf;
end
"""
rewrite(pg::ProbabilityGrammar, tokens::AbstractVector, into::AbstractVector)
Return the resulting array of words by replacing each token in 'tokens' with a random word.
"""
function rewrite(pg::ProbabilityGrammar, tokens::AbstractVector, into::AbstractVector)
local p::Float64;
for token in tokens
if (token in keys(pg.rules))
local non_terminal_symbols::AbstractVector;
non_terminal_symbols, p = weighted_choice(pg.rules[token]);
into[2] = into[2] * p;
rewrite(pg, non_terminal_symbols, into);
elseif (token in keys(pg.lexicon))
local terminal_symbol::String;
terminal_symbol, p = weighted_choice(pg.lexicon[token]);
push!(into[1], terminal_symbol);
into[2] = into[2] * p;
else
push!(into[1], token);
end
end
return into;
end
"""
generate_random_sentence(pg::ProbabilityGrammar, categories::String)
generate_random_sentence(pg::ProbabilityGrammar)
Return a randomly generated sentence as a String by using the given categories 'categories'.
"""
function generate_random_sentence(pg::ProbabilityGrammar, categories::String)
local rewritten_sentence::AbstractVector;
local p::Float64;
rewritten_sentence, p = rewrite(pg, split(categories), [[], 1.0]);
return (join(rewritten_sentence, " "), p);
end
function generate_random_sentence(pg::ProbabilityGrammar)
return generate_random_sentence(pg, "S");
end
# Create a submodule to contain the following example grammars.
# These grammars are defined in the module path "aimajulia.nlp".
module nlp
using ..aimajulia;
epsilon_0 = Grammar("ε_0",
Rules( # Grammar for ε_0 (Fig. 22.4 2nd edition)
["S"=>"NP VP | S Conjunction S",
"NP"=>"Pronoun | Name | Noun | Article Noun | Digit Digit | NP PP | NP RelClause",
"VP"=>"Verb | VP NP | VP Adjective | VP PP | VP Adverb",
"PP"=>"Preposition NP",
"RelClause"=>"That VP"]),
Lexicon( # Lexicon for ε_0 (Fig. 22.3 2nd edition)
["Noun"=>"stench | breeze | glitter | nothing | wumpus | pit | pits | gold | east",
"Verb"=>"is | see | smell | shoot | fell | stinks | go | grab | carry | kill | turn | feel",
"Adjective"=>"right | left | east | south | back | smelly",
"Adverb"=>"here | there | nearby | ahead | right | left | east | south | back",
"Pronoun"=>"me | you | I | it",
"Name"=>"John | Mary | Boston | Aristotle",
"Article"=>"the | a | an",
"Preposition"=>"to | in | on | near",
"Conjunction"=>"and | or | but",
"Digit"=>"0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9"
]));
# An example grammar for testing
epsilon_ = Grammar("ε_",
Rules(["S"=>"NP VP",
"NP"=>"Art N | Pronoun",
"VP"=>"V NP"]),
Lexicon(["Art"=>"the | a",
"N"=>"man | woman | table | shoelace | saw",
"Pronoun"=>"I | you | it",
"V"=>"saw | liked | feel"]));
# An example grammar for testing
epsilon_np_ = Grammar("ε_NP_",
Rules(["NP"=>"Adj NP | N"]),
Lexicon(["Adj"=>"happy | handsome | hairy", "N"=>"man"]));
# ε_probability is a probabilistic grammar found in the Python notebook.
# The rules (Fig 23.2 3rd edition) for the language 'ε_probability' use the probabilities from the Python notebook.
epsilon_probability = ProbabilityGrammar("ε_probability",
ProbabilityRules(["S"=>"NP VP [0.6] | S Conjunction S [0.4]",
"NP"=>("Pronoun [0.2] | Name [0.05] | Noun [0.2] | "*
"Article Noun [0.15] | Article Adjs Noun [0.1] | "*
"Digit [0.05] | NP PP [0.15] | NP RelClause [0.1]"),
"VP"=>("Verb [0.3] | VP NP [0.2] | VP Adjective [0.25] | "*
"VP PP [0.15] | VP Adverb [0.1]"),
"Adjs"=>"Adjective [0.5] | Adjective Adjs [0.5]",
"PP"=>"Preposition NP [1]",
"RelClause"=>"RelPro VP [1]"]),
ProbabilityLexicon(["Verb"=>"is [0.5] | say [0.3] | are [0.2]",
"Noun"=>"robot [0.4] | sheep [0.4] | fence [0.2]",
"Adjective"=>"good [0.5] | new [0.2] | sad [0.3]",
"Adverb"=>"here [0.6] | lightly [0.1] | now [0.3]",
"Pronoun"=>"me [0.3] | you [0.4] | he [0.3]",
"RelPro"=>"that [0.5] | who [0.3] | which [0.2]",
"Name"=>"john [0.4] | mary [0.4] | peter [0.2]",
"Article"=>"the [0.5] | a [0.25] | an [0.25]",
"Preposition"=>"to [0.4] | in [0.3] | at [0.3]",
"Conjuction"=>"and [0.5] | or [0.2] | but [0.3]",
"Digit"=>"0 [0.35] | 1 [0.35] | 2 [0.3]"]));
epsilon_chomsky = Grammar("ε_chomsky",
Rules(["S"=>"NP VP",
"NP"=>"Article Noun | Adjective Noun",
"VP"=>"Verb NP | Verb Adjective"]),
Lexicon(["Article"=>"the | a | an",
"Noun"=>"robot | sheep | fence",
"Adjective"=>"good | new | sad",
"Verb"=>"is | say | are"]));
epsilon_probability_chomsky = ProbabilityGrammar("ε_probability_chomsky",
ProbabilityRules(["S"=>"NP VP [1]",
"NP"=>"Article Noun [0.6] | Adjective Noun [0.4]",
"VP"=>"Verb NP [0.5] | Verb Adjective [0.5]"]),
ProbabilityLexicon(["Article"=>"the [0.5] | a [0.25] | an [0.25]",
"Noun"=>"robot [0.4] | sheep [0.4] | fence [0.2]",
"Adjective"=>"good [0.5] | new [0.2] | sad [0.3]",
"Verb"=>"is [0.5] | say [0.3] | are [0.2]"]));
end
#=
Chart is a data structure used in the process of analyzing a string
of words to uncover the phrase structure.
=#
mutable struct Chart
trace::Bool
grammar::Grammar
chart::AbstractVector
function Chart(g::Grammar; trace::Bool= false)
return new(trace, g);
end
end
"""
parse_sentence(chart::Chart, words::String, categories::String)
parse_sentence(chart::Chart, words::String)
parse_sentence(chart::Chart, words::Array{String, 1}, categories::String)
parse_sentence(chart::Chart, words::Array{String, 1})
Return an array of parses given the sentence 'words' and categories 'categories'.
"""
function parse_sentence(chart::Chart, words::String, categories::String)
local words_array::Array{String, 1} = map(String, split(words));
return parse_sentence(chart, words_array, categories);
end
function parse_sentence(chart::Chart, words::String)
local words_array::Array{String, 1} = map(String, split(words));
return parse_sentence(chart, words_array, "S");
end
function parse_sentence(chart::Chart, words::Array{String, 1}, categories::String)
parse_words(chart, words, categories);
return collect([i, j, categories, found, []]
for (i, j, lhs, found, expects) in chart.chart[length(words)]);
end
function parse_sentence(chart::Chart, words::Array{String, 1})
return parse_sentence(chart, words, "S");
end
"""
parse_words(chart::Chart, words::Array{String, 1}, categories::String)
parse_words(chart::Chart, words::Array{String, 1})
Return an array of words given the array of 'words' and categories 'categories'.
"""
function parse_words(chart::Chart, words::Array{String, 1}, categories::String)
chart.chart = collect([] for i in 1:(length(words) + 1));
add_edge(chart, [1, 1, "S_", [], [categories]]);
for i in 1:length(words)
scanner(chart, i, words[i]);
end
return chart.chart;
end
function parse_words(chart::Chart, words::Array{String, 1})
return parse_words(chart, words, "S");
end
"""
add_edge(chart::Chart, edge::AbstractVector)
Add the given edge 'edge' to the chart 'chart'.
"""
function add_edge(chart::Chart, edge::AbstractVector)
local start_index::Int64;
local end_index::Int64;
local lhs::String;
local found::AbstractVector;
local expects::AbstractVector;
start_index, end_index, lhs, found, expects = edge;
if (!(edge in chart.chart[end_index]))
push!(chart.chart[end_index], edge);
if (chart.trace)
println("Chart: added ", edge);
end
if (length(expects) == 0)
extender(chart, edge);
else
predictor(chart, edge);
end
end
return nothing;
end
"""
scanner(chart::Chart, index::Int64, word::String)
Extend the edge for given index 'index' if the given word 'word' and its category are expected.
"""
function scanner(chart::Chart, index::Int64, word::String)
for (i, j, A, alpha, Bb) in chart.chart[index]
if ((length(Bb) != 0) && (is_category(chart.grammar, word, Bb[1])))
add_edge(chart, [i, (j + 1), A, vcat(alpha, [(Bb[1], word)]), Bb[2:end]]);
end
end
return nothing;
end
"""
predictor(chart::Chart, edge::AbstractVector)
Add edges with rules for 'B' that may help in extending the given edge 'edge'.
"""
function predictor(chart::Chart, edge::AbstractVector)
local i::Int64;
local j::Int64;
local A::String;
local alpha::AbstractVector;
local Bb::AbstractVector;
i, j, A, alpha, Bb = edge;
local B::String = Bb[1];
if (haskey(chart.grammar.rules, B))
for rhs in rewrites_for(chart.grammar, B)
add_edge(chart, [j, j, B, [], rhs]);
end
end
return nothing;
end
"""
extender(chart::Chart, edge::AbstractVector)
Extend whatever edge can be extended from the given edge 'edge' to the chart 'chart'.
"""
function extender(chart::Chart, edge::AbstractVector)
local m::Int64;
local n::Int64;
local B::String;
local found::AbstractVector;
local expects::AbstractVector;
m, n, B, found, expects = edge;
for (i, j, A, alpha, B1b) in chart.chart[m]
if ((length(B1b) != 0) && (B == B1b[1]))
add_edge(chart, [i, n, A, vcat(alpha, [edge]), B1b[2:end]]);
end
end
return nothing;
end
"""
cyk_parse(words::Array{String, 1}, grammar::ProbabilityGrammar)
Return the resulting table as a Dict by applying the CYK algorithm (Fig. 23.5) on the
given sequence of words 'words' and grammar 'grammar'.
"""
function cyk_parse(words::Array{String, 1}, grammar::ProbabilityGrammar)
local N::Int64 = length(words);
local P::Dict = Dict();
for (i, word) in enumerate(words)
for (X, p) in get!(grammar.categories, word, [])
P[(X, i, 1)] = p;
end
end
for l in 2:N
for start_index in 1:(N - l + 1)
for len1 in 1:(l - 1)
local len2::Int64 = l - len1;
for (X, Y, Z, p) in cnf_rules(grammar)
P[(X, start_index, l)] = max(get!(P, (X, start_index, l), 0.0),
(get!(P, (Y, start_index, len1), 0.0)
* get!(P, (Z, start_index + len1, len2), 0.0) * p));
end
end
end
end
return P;
end
#=
Page consists of an address, an array of inlinks, an array of outlinks, a hub score, and an authority score.
=#
mutable struct Page
address::String
inlinks::AbstractVector
outlinks::AbstractVector
hub::Float64
authority::Float64
function Page(address::String; inlinks::AbstractVector=[], outlinks::AbstractVector=[], hub::Int64=0, authority::Int64=0)
return new(address, inlinks, outlinks, hub, authority);
end
end
"""
load_page_html(addresses::AbstractVector)
Return a Dict of page content for the given URLs 'addresses' by downloading the HTML pages.
"""
function load_page_html(addresses::AbstractVector)
local content::Dict = Dict();
for address in addresses
local tmp_file::Tuple = mktemp(pwd());
close(tmp_file[2]);
rm(tmp_file[1]);
download(address, tmp_file[1]);
local raw_html::String = String(read(tmp_file[1]));
rm(tmp_file[1]);
local html::String = replace(raw_html, @r_str("<head>(.*)</head>", "s")=>"");
content[address] = html;
end
return content;
end
"""
determine_inlinks(page::Page, pages_index::Dict)
Return an Array of inlinks for the given page 'page'.
"""
function determine_inlinks(page::Page, pages_index::Dict)
local inlinks::AbstractVector = [];
for (address, index_page) in pages_index
if (page.address == index_page.address)
continue;
elseif (page.address in index_page.outlinks)
push!(inlinks, address);
end
end
return inlinks;
end
"""
find_outlinks(page::Page, pages_content::Dict)
find_outlinks(page::Page, pages_content::Dict, handle_urls::Function)
Return an Array of outlinks to other pages for the given page 'page'.
If the argument 'handle_urls' is given, the resulting array is returned
after the function is applied to the array of outlinks.
"""
function find_outlinks(page::Page, pages_content::Dict)
local urls::AbstractVector = Array{String, 1}();
for regex_m in eachmatch(r"href=['\"]?([^'\" >]+)", pages_content[page.address])
push!(urls, String(regex_m.captures[1]));
end
return urls;
end
function find_outlinks(page::Page, pages_content::Dict, handle_urls::Function)
local urls::AbstractVector = Array{String, 1}();
for regex_m in eachmatch(r"href=['\"]?([^'\" >]+)", pages_content[page.address])
push!(urls, String(regex_m.captures[1]));
end
return handle_urls(urls);
end
"""
init_pages(addresses::AbstractVector)
Return a Dict of pages from the given array of URLs 'addresses'.
"""
function init_pages(addresses::AbstractVector)
local pages::Dict = Dict();
for address in addresses
pages[address] = Page(address);
end
return pages;
end
"""
only_wikipedia_urls(urls::AbstractVector)
Return an Array of wikipedia URLs where relative wiki URLs are converted to their absolute URL.
"""
function only_wikipedia_urls(urls::AbstractVector)
local wiki_urls::AbstractVector = collect(url for url in urls if (startswith(url, "/wiki/")));
return collect(("https://en.wikipedia.org" * url) for url in wiki_urls);
end
"""
expand_pages(pages::Dict, pages_index::Dict)
Return the Dict of expanded pages by adding in every page that links to
or is linked from one of the relevant pages.
"""
function expand_pages(pages::Dict, pages_index::Dict)
local expanded::Dict = Dict();
for (address, page) in pages
if (!haskey(expanded, address))
expanded[address] = page;
end
for inlink in page.inlinks
if (!haskey(expanded, inlink))
expanded[inlink] = pages_index[inlink];
end
end
for outlink in page.outlinks
if (!haskey(expanded, outlink))
expanded[outlink] = pages_index[outlink];
end
end
end
return expanded;
end
"""
relevant_pages(query::String, pages_index::Dict, pages_content::Dict)
Return a Dict of pages that contain all of the query words in 'query'.
These pages are found by intersecting the hit lists of the query words.
"""
function relevant_pages(query::String, pages_index::Dict, pages_content::Dict)
local hit_intersection::Set = Set(collect(keys(pages_index)));
local query_words::AbstractVector = map(String, split(query));
for query_word in query_words
local hit_list::Set = Set();
for address in keys(pages_index)
if (occursin(lowercase(query_word), lowercase(pages_content[address])))
push!(hit_list, address);
end
end
hit_intersection = intersect(hit_intersection, hit_list);
end
return Dict(collect((address, pages_index[address]) for address in hit_intersection));
end
"""
normalize_pages(pages::Dict)
Divide the scores of each page in 'pages' by the sum of all the squares of all pages' scores
(separately for both the authority and hubs scores).
"""
function normalize_pages(pages::Dict)
local summed_hub::Float64 = sum(collect(page.hub^2 for page in values(pages)));
local summed_authority::Float64 = sum(collect(page.authority^2 for page in values(pages)));
local sqrt_summed_hub::Float64 = sqrt(summed_hub);
local sqrt_summed_authority::Float64 = sqrt(summed_authority);
for address in keys(pages)
pages[address].hub = pages[address].hub / sqrt_summed_hub;
pages[address].authority = pages[address].authority / sqrt_summed_authority;
end
return nothing;
end
#=
ConvergenceDetector contains the hub history and authorities history for a set of pages.
=#
struct ConvergenceDetector
hub_history::AbstractVector
authority_history::AbstractVector
function ConvergenceDetector()
return new([], []);
end
end
"""
detect_convergence(cd::ConvergenceDetector, pages_index::Dict)
Return a boolean indicating if both the 'hub' and 'authority' values of the pages
in 'pages_index' have converged.
"""
function detect_convergence(cd::ConvergenceDetector, pages_index::Dict)
local current_hubs::AbstractVector = collect(page.hub for page in values(pages_index));
local current_authorities::AbstractVector = collect(page.authority for page in values(pages_index));
if (length(cd.hub_history) != 0)
local diffs_hub::AbstractVector = collect(abs(x - y) for (x, y) in zip(current_hubs, cd.hub_history[end]));
local diffs_authority::AbstractVector = collect(abs(x - y) for (x, y) in zip(current_authorities, cd.authority_history[end]));
local average_delta_hub::Float64 = sum(diffs_hub)/length(pages_index);
local average_delta_authority::Float64 = sum(diffs_authority)/length(pages_index);
if ((average_delta_hub < 0.01) && (average_delta_authority < 0.01))
return true;
end
end
if (length(cd.hub_history) > 2)
deleteat!(cd.hub_history, 1);
deleteat!(cd.authority_history, 1);
end
push!(cd.hub_history, current_hubs);
push!(cd.authority_history, current_authorities);
return false;
end
"""
get_inlinks(page::Page, pages_index::Dict)
Return an Array of addresses where each address is in both the page's inlinks
and index of pages 'pages_index'.
"""
function get_inlinks(page::Page, pages_index::Dict)
if (length(page.inlinks) == 0)
page.inlinks = determine_inlinks(page, pages_index);
end
return collect(address for (address, p) in pages_index if (address in page.inlinks));
end
"""
get_outlinks(page::Page, pages_index::Dict, pages_content::Dict)
Return an Array of addresses where each address is in both the page's outlinks
and index of pages 'pages_index'.
"""
function get_outlinks(page::Page, pages_index::Dict, pages_content::Dict)
if (length(page.outlinks) == 0)
page.outlinks = find_outlinks(page, pages_content);
end
return collect(address for (address, p) in pages_index if (address in page.outlinks));
end
"""
HITS(query::String, pages_index::Dict, pages_content::Dict)
Return the computed hubs and authorities with respect to the query as a Dict by using the
HITS algorithm (Fig. 22.1) to the given query 'query', index of pages 'pages_index',
and the content of the pages 'pages_content'.
"""
function HITS(query::String, pages_index::Dict, pages_content::Dict)
local pages::Dict = expand_pages(relevant_pages(query, pages_index, pages_content), pages_index);
for p in values(pages)
p.authority = 1.0;
p.hub = 1.0;
end
local convergence::ConvergenceDetector = ConvergenceDetector();
while (detect_convergence(convergence, pages_index))
local authority::Dict = Dict(collect(Pair(p, pages[p].authority) for p in pages));
local hub::Dict = Dict(collect(Pair(p, pages[p].hub) for p in pages));
for p in values(pages)
p.authority = sum(hub[x] for x in get_inlinks(p, pages_index));
p.hub = sum(authority[x] for x in get_outlinks(p, pages_index, pages_content));
end
normalize_pages(pages);
end
return pages;
end