treat 0.1.1

data/lib/treat/extractors/statistics/transition_probability.rb

@@ -0,0 +1,53 @@
+module Treat
+  module Extractors
+    module Statistics
+      class TransitionProbability
+
+        # Find the transition probability.
+        def self.statistics(entity, options={})
+          tm = options[:transition_matrix]
+          features = tm.keys
+          score = 0
+          count = 0
+
+          relationships = [:parent, :left, :right, :children]
+          
+          features.each do |f1|
+            v1 = entity.send(f1)
+            features.each do |f2|
+              next unless tm[f1][v1]
+              
+              relationships.each do |relationship|
+                relatives = target.send(relationship)
+                relatives = [relatives] unless relatives.is_a? Array
+                relatives.each do |relative|
+                  next if relative.nil? || !relative.has?(f2)
+                  v2 = relative.send(f2)
+                  if tm[f1][v1][relationship][f2][v2]
+                    score += tm[f1][v1][relationship][f2][v2]
+                    count += 1
+                  end
+                end
+              end
+
+              entity.edges.each do |id, edge|
+                s = entity.ancestor_with_type :sentence
+                if s
+                  x = s.find(id)
+                  next unless h.has?(f2)
+                  v2 = x.send(f2)
+                  if tm[f1][v1][:edge][f2][v2]
+                    score += tm[f1][v1][:edge][f2][v2]
+                    count += 1
+                  end
+                end
+              end
+
+            end
+          end
+          score.to_f/count.to_f
+        end
+      end
+    end
+  end
+end

data/lib/treat/extractors/time/chronic.rb

@@ -0,0 +1,12 @@
+module Treat
+  module Extractors
+    module Time
+      class Chronic
+        silently { require 'chronic' }
+        def self.time(entity, options = {})
+          silently { ::Chronic.parse(entity.to_s, {:guess => true}) }
+        end
+      end
+    end
+  end
+end

data/lib/treat/extractors/time/native.rb

@@ -0,0 +1,12 @@
+module Treat
+  module Extractors
+    module Time
+      module Native
+        require 'date'
+        def self.time(entity, options = {})
+          ::DateTime.parse(entity.to_s)
+        end
+      end
+    end
+  end
+end

data/lib/treat/extractors/time/nickel.rb

@@ -0,0 +1,45 @@
+module Treat
+  module Extractors
+    module Time
+=begin
+      Annotations
+
+      Type examples
+      
+      single "lunch with megan tomorrow at noon"
+      daily "Art exhibit until March 1st"
+      weekly "math class every wed from 8-11am"
+      daymonthly "open bar at joes the first friday of every month"
+      datemonthly "pay credit card bill on the 22nd of each month"
+
+=end
+      module Nickel
+        require 'date'
+        silently { require 'nickel' }
+        def self.time(entity, options = {})
+          n = silently { ::Nickel.parse(entity.to_s) }
+          occ = n.occurrences[0]
+          # Find the words..
+          rec = occ.type.to_s.gsub('single', 'once').intern
+          entity.set :time_recurrence, rec
+          interval = occ.interval ? occ.interval.intern : :none
+          entity.set :time_recurrence_interval, interval
+          
+          s = [occ.start_date, occ.start_time]
+          ds = [s[0].year, s[0].month, s[0].day] if s[0]
+          ts = [s[1].hour, s[1].min, s[1].sec] if s[1]
+          
+          e = [occ.end_date, occ.end_time]
+          de = [e[0].year, e[0].month, e[0].day] if e[0]
+          te = [e[1].hour, e[1].min, e[1].sec] if e[1]
+          
+          entity.set :start_time, ::DateTime.civil(*ds, *ts) if ds
+          entity.set :end_time, ::DateTime.civil(*de, *te) if de
+          
+          entity.start_time
+        end
+      end
+    end
+  end
+end
+

data/lib/treat/extractors/topic_words/lda.rb

@@ -0,0 +1,71 @@
+module Treat
+  module Extractors
+    module TopicWords
+      # An adapter for the 'lda-ruby' gem, which clusters
+      # documents into topics based on Latent Dirichlet
+      # Allocation.
+      #
+      # Original paper:
+      # Blei, David M., Ng, Andrew Y., and Jordan, Michael
+      # I. 2003. Latent dirichlet allocation. Journal of
+      # Machine Learning Research. 3 (Mar. 2003), 993-1022.
+      class LDA
+        # Require the lda-ruby gem.
+        silently { require 'lda-ruby' }
+        # Monkey patch the TextCorpus class to call it without
+        # having to create any files.
+        Lda::TextCorpus.class_eval do
+          # Ruby, Y U NO SHUT UP!
+          silently { undef :initialize }
+          # Redefine initialize to take in an array of texts.
+          def initialize(texts)
+            super(nil)
+            texts.each do |text|
+              add_document(Lda::TextDocument.new(self, text))
+            end
+          end
+        end
+        def self.topic_words(collection, options = {})
+          # Set the options
+          options[:words_per_topic] ||= 10
+          options[:topics] ||= 20
+          options[:iterations] ||= 20
+
+          # Create a corpus with the collection
+          texts = collection.texts.collect do |t| 
+            t.to_s.encode("UTF-8", :invalid => :replace, :undef => :replace, :replace => "?")
+          end
+          corpus = Lda::TextCorpus.new(texts)
+
+          # Create an Lda object for training
+          lda = Lda::Lda.new(corpus)
+          lda.num_topics = options[:topics]
+          lda.max_iter = options[:iterations]
+          # Run the EM algorithm using random starting points
+          silence_streams(STDOUT, STDERR) { lda.em('random') }
+          
+          # Load the vocabulary.
+          if options[:vocabulary]
+            lda.load_vocabulary(options[:vocabulary])
+          end
+          
+          # Get the topic words and annotate the text.
+          topic_words = lda.top_words(options[:words_per_topic])
+          
+          topic_words.each do |i, words|
+            collection.each_word do |word|
+              if words.include?(word)
+                word.set :is_topic_word?, true
+                word.set :topic_id, i
+              else
+                word.set :is_topic_word?, false
+              end
+            end
+          end
+          
+          topic_words
+        end
+      end
+    end
+  end
+end

data/lib/treat/extractors/topic_words/lda/data.dat

@@ -0,0 +1,46 @@
+170 219:2 256:2 389:1 257:1 292:2 143:1 181:14 107:1 1871:1 241:1 104:14 120:1 183:1 175:1 267:1 274:1 245:2 884:1 996:1 202:1 149:2 1264:1 182:1 275:1 2015:1 212:1 1292:1 367:1 142:1 221:2 1328:2 265:1 152:1 494:1 186:3 168:1 216:1 1184:1 284:2 276:1 151:1 164:1 290:1 249:1 150:1 141:1 125:2 1281:2 1953:1 196:1 112:3 281:1 314:1 934:2 286:2 134:1 148:6 225:1 114:2 211:1 147:1 300:1 303:1 266:1 201:1 191:1 124:8 244:1 209:1 170:2 517:1 105:1 248:1 7:1 155:2 675:1 246:1 571:1 224:1 285:1 289:1 119:3 230:4 262:1 2342:1 159:1 205:1 217:1 195:2 944:1 220:2 106:1 277:1 137:1 135:2 1458:1 118:9 2368:1 199:2 1798:1 242:1 239:2 127:3 264:3 2129:1 2076:1 103:1 154:2 238:1 102:5 108:1 197:1 268:1 184:2 19:2 255:1 430:1 173:1 309:1 138:2 261:1 272:1 5:2 128:3 1145:7 192:1 129:15 145:1 236:1 121:8 193:4 1356:1 140:1 840:3 171:2 172:2 721:1 935:1 160:2 185:1 207:1 1086:1 126:1 1132:1 130:5 157:1 271:2 100:4 132:1 200:1 188:6 204:1 109:5 153:1 158:1 304:1 208:1 146:3 110:2 218:1 2079:1 868:1 210:1 557:3 227:1 282:2 247:2 165:1 213:1 215:1
+150 219:1 389:1 257:4 181:2 104:7 120:1 427:1 416:1 274:1 1517:1 310:4 373:1 245:4 352:3 479:1 1661:1 400:1 2015:1 401:1 318:1 1887:1 367:1 381:1 413:1 221:1 993:1 186:5 1543:1 358:1 334:1 1184:1 276:1 421:2 340:1 290:1 422:3 368:1 125:1 112:2 314:1 333:1 286:2 2157:1 134:2 424:2 148:3 362:1 225:1 300:1 147:1 382:2 311:4 303:1 361:2 330:1 170:1 124:5 351:4 517:1 7:1 326:1 675:1 343:1 402:1 399:3 571:4 119:2 97:1 411:1 205:1 345:1 360:2 220:1 137:2 1058:1 323:1 372:1 1798:1 375:1 1349:1 2129:1 420:1 154:1 19:2 430:1 398:1 309:1 429:1 434:2 325:1 384:1 426:1 261:1 337:2 412:1 366:1 5:2 316:1 1145:1 357:1 129:24 1727:1 121:3 403:1 433:1 193:3 339:1 356:1 317:1 380:1 1721:2 741:1 484:1 880:1 338:1 324:3 185:1 1086:1 341:2 2051:1 130:5 387:1 423:1 347:2 1610:1 132:1 350:1 109:4 320:4 331:1 406:1 394:1 431:1 393:2 392:1 344:2 304:1 391:1 354:2 335:1 432:1 329:1 2219:1 557:2 385:1 397:1 414:1 379:1 313:1 1980:1
+95 186:1 512:2 242:1 448:5 449:5 127:1 494:1 276:3 151:1 497:1 507:1 205:1 286:1 480:2 460:1 473:1 324:1 489:1 517:1 461:1 1158:1 471:1 2087:1 493:3 437:2 1546:2 271:2 146:1 479:2 1661:2 544:2 466:1 191:2 514:1 1034:1 519:1 129:9 481:2 94:2 499:1 485:1 508:1 442:5 571:3 463:2 320:3 498:2 329:1 1621:2 1481:2 1154:1 290:1 109:3 225:2 515:1 447:2 282:1 439:2 464:1 496:1 130:6 441:1 422:2 505:1 504:1 450:1 495:1 484:1 121:6 41:1 503:1 467:1 132:2 1798:1 500:1 217:3 257:2 1912:1 440:4 1472:1 478:1 476:1 104:12 1302:1 325:1 513:2 453:1 32:2 193:1 743:1 475:2 516:1 1557:2 1078:1 474:1
+249 102:6 1981:1 689:1 718:2 1145:4 745:1 804:1 631:1 507:1 588:1 683:1 1244:1 148:2 34:1 325:2 736:1 585:2 677:1 580:1 685:1 741:1 337:2 551:1 590:1 682:1 574:6 596:1 1734:4 594:1 598:1 181:2 603:3 1075:2 154:3 552:1 1557:2 744:1 217:1 236:1 621:1 700:2 249:3 1481:2 601:1 282:1 549:1 713:1 281:1 2071:1 537:1 112:1 728:1 264:1 670:2 602:1 274:2 684:1 1278:1 389:1 91:1 512:2 673:1 721:3 1914:1 668:2 124:1 610:1 527:1 121:11 515:1 629:2 699:1 559:1 698:1 560:1 397:1 719:1 324:1 666:1 525:4 1605:3 168:5 411:2 276:13 320:4 625:1 611:1 692:3 185:1 722:3 563:1 657:1 747:1 647:1 261:2 690:2 751:1 679:1 664:2 717:1 681:1 701:2 1546:4 529:1 406:1 526:4 450:1 442:1 88:1 696:2 5:17 662:1 503:1 129:22 1658:1 286:5 553:1 640:1 104:30 615:1 354:6 565:1 544:4 746:1 714:1 1034:1 589:1 1621:2 605:1 612:1 170:4 125:1 523:1 607:3 557:8 540:1 644:1 2001:1 743:2 578:1 191:1 675:1 740:1 697:1 146:2 41:4 522:1 592:1 2148:2 730:1 634:1 547:7 192:3 362:1 2067:1 151:3 731:2 1158:1 632:1 205:2 516:1 723:1 619:1 649:1 1661:1 277:3 137:2 733:1 1302:1 734:6 429:1 2368:4 360:2 604:1 1504:1 193:5 2048:1 329:3 597:1 659:1 109:8 630:1 725:1 708:1 749:1 514:1 593:1 691:1 724:1 568:1 618:3 686:1 958:1 186:9 614:3 581:1 738:1 2223:1 130:20 19:10 934:3 550:1 257:2 528:5 726:1 595:1 687:1 1094:1 616:1 571:8 498:2 655:1 513:2 586:1 660:2 517:3 225:2 573:1 132:2 608:4 1251:1 575:1 290:4 1154:1 32:3 709:1 914:1 530:1 620:1 652:2 555:5 548:1 729:2 188:1 637:1 300:5 583:4 7:2 245:1 171:1 479:4 1239:1 422:4 1592:1 2214:5 645:1 543:1 688:2 626:2
+141 795:1 256:2 773:1 573:1 257:1 753:1 2001:1 292:1 181:7 1278:1 2231:1 104:5 183:1 802:1 1481:2 245:2 453:1 677:2 479:2 1546:3 781:1 1096:1 804:1 821:1 780:1 186:3 782:1 1347:1 168:1 284:1 151:1 276:2 760:1 767:1 290:2 1621:2 812:1 824:1 596:1 801:1 618:2 112:1 574:3 788:1 934:2 286:3 796:1 148:1 544:2 513:4 820:1 300:2 790:1 124:1 754:2 1866:1 777:1 517:2 776:1 7:1 799:1 816:1 571:4 762:1 285:1 752:1 810:2 1034:1 230:1 769:1 809:1 205:1 751:4 217:3 756:1 818:1 220:1 768:1 464:1 498:1 823:1 1158:1 199:1 743:1 1228:1 803:1 2129:1 239:1 154:1 102:3 1981:2 86:1 514:1 19:2 815:1 471:1 450:1 778:1 325:2 775:1 5:5 1145:2 1534:1 129:14 2253:1 121:7 193:3 797:1 793:1 787:1 317:1 526:1 772:1 172:1 516:1 324:1 512:2 17:1 2214:2 783:1 1239:1 770:1 130:4 515:1 271:1 1575:1 1357:1 766:1 109:3 320:1 188:2 1085:1 817:1 1302:1 329:1 282:1 1094:1 786:1 1557:2 32:1 792:1
+224 102:9 835:1 689:1 850:1 1145:5 890:1 804:5 507:1 1544:2 1244:1 1049:1 874:1 325:1 148:8 241:1 900:1 960:1 989:1 987:1 142:1 866:1 833:1 996:1 219:1 574:7 1734:1 2116:2 909:2 181:18 603:1 1347:2 1557:2 217:4 1895:1 957:1 955:1 845:1 951:1 1481:2 282:3 975:1 919:1 898:1 988:1 281:2 265:1 844:1 220:1 112:2 264:3 829:2 979:2 274:2 1278:1 917:1 980:2 950:1 982:1 952:1 175:2 512:2 920:1 1132:3 515:1 121:11 947:1 345:1 886:1 560:1 888:1 324:1 160:1 853:1 168:1 494:1 990:1 1830:1 276:3 256:3 932:5 519:1 963:1 199:1 970:2 1498:1 426:1 881:1 918:1 165:1 747:1 837:2 896:1 1703:1 880:1 832:1 840:2 701:1 882:2 1546:2 885:1 864:1 846:1 875:1 2229:1 413:1 1185:1 172:1 129:15 286:3 956:2 615:1 897:1 944:1 104:11 344:1 935:1 544:2 910:1 978:1 907:1 924:1 1034:1 843:1 2015:1 903:1 2191:1 901:1 938:1 170:1 1621:4 923:1 484:1 883:1 134:2 379:1 849:3 865:2 1523:1 927:1 743:2 191:3 948:1 697:1 1472:1 941:1 1354:1 933:1 127:2 151:3 210:1 1158:1 1000:1 516:1 831:6 993:1 367:1 1302:1 2368:1 137:1 1281:1 1505:1 1661:2 193:4 830:3 983:1 109:8 962:1 922:1 514:1 994:1 593:2 946:1 999:1 618:1 958:1 868:5 738:1 891:1 186:3 130:13 965:2 934:4 271:1 1014:1 257:1 1873:1 352:1 571:7 2089:2 498:1 513:2 939:1 828:1 517:2 225:3 916:1 132:2 964:1 902:1 926:1 230:3 290:12 969:1 412:1 914:1 530:1 876:2 884:3 859:1 555:2 729:1 221:1 984:1 1078:1 894:2 188:3 1820:1 227:3 245:2 239:1 422:1 1328:2 479:3 937:1
+66 1038:1 818:1 81:2 743:1 453:2 1017:1 80:1 193:3 264:1 866:1 241:1 130:1 129:3 1125:1 1024:1 934:1 175:1 809:2 1033:1 225:1 154:1 1493:1 1007:1 2148:1 741:1 2219:1 1356:1 1027:1 1004:2 1015:1 121:3 325:1 274:1 109:1 104:6 191:1 1016:1 1031:1 955:1 406:1 1020:1 1010:1 430:1 574:3 290:1 1028:1 1029:1 220:1 903:1 188:1 261:1 517:3 1035:1 1026:1 1145:2 1034:1 1021:1 275:1 102:1 362:1 1022:1 282:1 965:1 2334:1 1014:1 257:1
+56 145:1 19:1 1672:1 1055:1 1050:1 193:2 1059:1 1042:1 130:4 170:1 129:6 1069:2 682:1 1043:2 1062:1 1051:1 1078:1 876:1 225:1 1053:1 1065:1 2219:1 1046:1 1961:1 1074:1 2352:1 1071:1 1068:1 1049:1 121:2 1351:8 1075:1 147:1 2214:1 109:2 1105:1 367:1 1056:1 168:1 1066:1 1077:1 329:1 422:1 1058:1 261:1 5:4 275:1 276:1 1072:1 102:1 958:1 286:2 125:1 1045:1 186:1 257:1
+34 19:2 1083:2 1082:2 112:1 129:1 1043:1 738:1 594:2 1085:1 2150:1 1094:1 32:2 121:3 1351:4 217:3 2214:1 1727:1 104:3 1084:1 1086:3 1895:1 5:1 1328:1 102:1 2132:1 615:1 1091:1 1108:1 286:1 1088:1 722:1 475:1 1090:1 227:1
+50 1118:1 453:1 1110:1 1117:1 271:1 130:1 170:1 129:1 882:2 1100:1 738:1 1111:1 154:1 1007:1 1109:1 956:1 728:1 2219:1 426:1 1101:1 1313:2 1099:1 121:1 1351:2 1105:3 104:4 1097:1 571:1 1096:1 1113:1 2008:1 1098:1 640:1 574:1 290:1 1106:1 165:1 220:1 1058:1 1034:1 1115:1 102:2 1119:1 1108:1 286:1 186:1 909:1 514:1 1103:1 2116:1
+51 1144:1 19:2 1126:1 1122:1 1017:1 581:1 786:1 1142:1 193:1 1141:1 112:1 130:2 129:1 1125:1 142:1 1143:1 2364:1 738:1 1129:2 1130:1 157:1 1127:2 1136:2 809:1 225:1 1472:3 1138:1 119:2 1140:1 121:3 1351:4 209:1 104:2 1123:1 571:2 148:1 1131:1 1098:1 290:1 1133:1 132:1 1146:1 5:4 1145:2 199:1 729:3 1132:1 1147:1 580:1 257:1 1135:1
+121 256:1 257:1 181:5 1154:1 611:2 910:2 104:7 274:1 1481:2 1078:2 453:2 1472:1 1185:2 1265:1 1171:1 1192:1 1169:1 1661:1 479:1 1546:2 1220:1 956:1 1182:1 1136:5 142:1 186:3 168:3 1153:2 1184:1 1210:1 151:1 555:2 249:1 422:2 1207:1 125:1 1621:4 876:2 1159:1 1181:1 574:4 1189:1 286:5 2101:1 513:2 148:3 1141:1 544:2 300:2 191:1 1254:1 1389:1 517:2 1172:1 1201:1 1165:1 571:3 2281:1 2132:2 924:3 217:1 1177:1 1285:1 205:1 277:1 498:1 1158:3 1195:1 743:1 1798:2 1208:1 1219:1 537:1 514:1 19:6 932:1 975:1 429:1 325:2 1164:2 5:6 1082:1 1145:1 1151:1 1083:1 701:1 129:12 734:2 193:4 41:4 121:4 565:2 2071:1 516:2 1197:4 324:1 512:2 962:1 130:8 515:1 1515:2 583:1 594:1 132:2 109:2 320:1 188:2 344:1 1211:1 1180:1 354:1 1135:1 1302:1 1162:1 329:1 557:5 1218:1 282:1 1557:2 32:3 519:1
+120 1258:1 257:1 1154:1 1241:1 603:1 1246:1 1113:1 104:4 1235:1 1481:2 1078:1 1472:1 1265:1 746:1 1264:1 1661:1 479:1 547:1 1546:2 1236:1 1257:1 1266:1 1249:1 1263:1 2087:2 956:1 1273:1 186:1 1256:1 626:1 168:2 216:2 1251:1 1269:1 151:1 276:3 249:4 1621:2 574:3 286:2 513:2 148:1 544:2 300:2 1262:2 559:1 1254:1 1767:1 495:1 105:1 517:1 571:2 499:1 762:1 810:3 522:1 119:2 540:1 217:1 1105:1 745:1 205:2 277:2 498:1 1158:1 199:2 1226:1 743:2 1228:1 1270:1 500:1 1255:1 514:1 2048:3 19:6 430:1 1245:1 450:2 2034:1 722:2 325:2 5:6 1082:1 1083:1 129:9 1274:1 193:6 41:2 121:5 2008:1 70:1 503:1 516:2 721:1 324:1 512:2 744:1 2214:3 1225:1 962:1 1239:1 130:6 515:1 980:1 583:1 204:1 1238:1 907:1 146:1 1252:1 1302:1 335:1 329:1 557:2 1869:1 282:1 1244:1 1557:2 32:1 519:1
+95 186:1 583:1 512:2 276:2 151:1 507:1 286:1 1292:1 1284:1 804:1 517:1 1158:1 241:1 1283:1 1281:1 1590:1 188:1 1546:2 1649:1 1238:1 762:1 1285:1 271:1 1747:1 581:1 19:3 932:1 479:2 544:2 191:1 514:1 5:4 1291:1 603:2 129:8 1141:2 1423:2 868:1 159:1 1303:1 1280:1 700:1 2001:1 571:1 102:1 1255:1 320:2 2214:1 498:1 329:1 1621:2 1481:2 1154:2 109:1 515:1 282:1 464:1 130:3 809:1 1534:1 119:1 300:1 450:1 1349:1 903:1 1278:2 192:1 1239:2 121:6 1624:2 1279:1 766:1 41:1 1301:1 1300:2 217:3 1286:1 1305:1 557:1 389:1 104:10 962:1 1302:2 2310:1 325:1 1228:1 513:2 32:2 8:1 193:3 743:1 181:3 1296:1 516:1 1557:2
+153 1332:3 1315:4 1317:5 256:3 389:2 1334:1 257:3 181:10 1296:1 910:1 1364:1 104:12 1481:2 175:3 373:1 1354:1 1265:2 1078:1 453:1 1320:1 1472:1 1325:1 479:2 1661:1 1546:2 1292:2 1376:1 1326:1 1378:1 804:2 221:1 1380:1 1328:2 1333:1 1347:4 168:1 939:1 688:1 729:1 151:1 276:3 1344:1 1016:1 290:9 1621:2 1281:1 1313:1 618:1 574:1 1384:1 934:4 286:2 134:2 796:1 148:4 544:2 513:2 225:2 1350:1 300:2 1327:1 170:1 754:2 517:1 1360:2 1051:2 1358:1 1385:5 816:1 571:3 1283:1 285:2 1351:1 522:2 15:1 119:1 345:2 1105:1 220:1 1340:1 2368:1 498:1 1158:1 199:1 264:1 743:1 1349:2 882:1 2129:1 239:1 154:1 102:10 955:1 514:1 19:6 932:3 450:4 1316:1 325:1 1335:1 261:1 1311:1 412:1 5:8 1534:2 1359:1 866:1 1339:1 1145:7 129:11 121:11 193:11 1346:1 1356:1 507:1 1373:1 741:2 772:1 172:1 850:1 516:1 1331:1 324:1 512:2 1368:1 1068:2 2214:4 1086:1 962:5 130:4 515:1 271:1 1348:1 1357:1 188:1 109:4 344:1 391:1 1302:1 1318:1 588:1 329:1 1094:3 227:1 282:1 894:1 1362:1 1557:2 864:1 519:1 1308:5 560:1 1355:1
+117 256:4 2147:1 181:10 978:1 1392:3 2158:1 104:6 175:1 1481:2 373:1 1265:1 746:1 245:5 182:1 400:2 479:1 1546:3 367:1 804:1 1347:1 168:1 186:2 729:1 151:2 276:2 1422:1 1410:1 249:3 1747:3 598:1 1621:2 530:1 574:3 1406:1 314:1 934:4 286:6 513:2 74:1 148:2 1141:1 544:2 300:1 1387:2 170:1 1411:1 1389:1 517:2 7:1 571:1 1402:2 762:2 119:1 810:4 1390:4 277:1 498:1 1404:1 1158:1 199:1 743:1 1228:2 127:4 154:1 500:1 525:1 102:1 514:1 19:6 430:1 932:2 450:1 722:1 1423:2 325:1 261:1 1388:4 5:5 1082:1 1401:1 1083:1 129:13 1396:1 193:4 926:1 41:1 121:9 1424:1 317:1 516:1 721:1 741:1 512:2 2214:1 962:1 130:7 515:1 980:1 583:1 132:1 1357:1 109:4 320:3 1417:2 8:1 1302:1 1416:1 282:1 329:1 868:2 1244:1 165:1 1557:2 32:1 786:1 1425:1 1355:1
+104 186:1 394:1 1453:1 512:2 818:1 108:3 151:1 662:1 1840:1 1427:1 1438:1 1452:2 286:1 199:1 924:1 220:1 1434:1 324:1 517:2 1244:2 1158:2 160:2 1430:4 241:1 1283:1 865:1 188:1 60:1 1546:3 19:1 479:1 1661:1 544:2 191:1 514:1 5:1 519:1 910:2 129:9 1123:2 1451:1 1433:1 1436:2 2129:1 571:3 574:1 320:1 1441:1 498:1 1444:1 964:1 1621:4 142:1 1481:2 110:1 290:1 109:6 1389:1 515:1 896:1 282:1 464:1 130:3 422:1 148:1 58:1 274:1 119:1 375:1 1145:2 1437:1 121:5 1162:1 227:1 1447:1 1798:1 217:2 840:1 1445:2 184:1 1435:1 557:1 1472:1 145:1 1456:1 104:13 962:1 325:1 1302:1 7:1 594:1 513:2 453:1 1449:1 193:2 1458:1 1446:4 743:1 1159:1 975:1 181:2 516:1 1557:2 1078:1
+89 692:1 1473:1 2347:1 168:6 2191:1 1488:1 512:2 2261:1 1592:1 276:1 151:1 1478:5 286:2 1469:1 1493:1 1226:1 804:1 2267:1 517:1 1158:1 1495:1 188:1 1546:3 762:1 146:1 19:2 479:1 544:2 514:1 5:3 910:1 129:11 2015:1 11:1 1094:1 499:1 729:1 696:1 1491:1 1498:1 574:3 2008:1 1490:1 571:2 102:1 2214:1 498:1 1621:2 1481:3 125:1 1389:1 1482:1 515:1 956:1 282:1 1480:1 130:4 2371:1 148:1 119:2 300:1 450:2 1475:1 1476:1 767:1 121:1 1279:2 766:3 1470:4 744:1 467:1 1497:1 1468:1 368:1 1472:2 145:2 104:11 1302:1 1113:1 325:1 1228:1 1486:1 513:2 193:5 743:1 1467:1 516:1 1557:2 1484:1
+163 257:2 1672:1 181:1 1241:1 1523:1 241:1 104:14 1541:1 1481:2 1246:1 1517:1 1078:1 1568:1 902:1 1472:1 677:1 1325:1 479:1 1661:1 1567:1 2021:1 1564:1 1546:3 1249:1 1923:1 1540:2 804:1 956:1 381:1 142:1 1505:1 413:1 780:1 186:1 782:1 1503:1 2229:1 1543:1 1333:1 1524:1 168:2 1184:1 939:1 688:2 729:4 151:1 276:9 249:1 1621:2 125:2 1516:1 80:1 618:2 530:1 1544:2 574:1 934:2 286:6 1506:1 544:2 513:2 1566:1 1547:1 1303:1 300:3 681:1 682:1 1527:1 191:1 754:1 946:1 1528:2 170:1 124:1 1532:1 1486:1 1480:1 1509:2 517:1 675:1 571:2 762:4 810:2 1501:1 119:1 11:1 809:1 205:2 751:1 1511:1 818:1 1571:1 1226:1 498:1 1520:1 1158:1 199:2 1798:1 264:2 743:1 1228:4 1549:1 2129:1 127:1 154:1 1533:1 1799:1 102:1 2130:1 514:2 19:3 1538:1 450:1 722:1 1423:1 325:1 337:1 5:8 1529:1 1534:3 1145:3 129:16 1569:1 145:1 121:9 193:3 1510:1 1502:2 1560:1 772:1 850:1 324:1 516:1 1559:1 512:2 17:1 185:1 2214:4 1539:1 130:9 515:1 271:1 1515:1 132:1 109:4 1554:1 1252:1 1504:1 158:1 1482:1 146:2 581:1 1302:1 335:1 329:1 1094:2 1305:1 1869:1 282:2 560:1 1557:3 1507:1 519:1
+112 186:2 1583:2 1632:1 127:2 512:2 1587:4 1589:1 1547:1 1592:1 625:1 1575:1 276:3 151:1 1594:1 507:1 205:2 286:3 1540:4 199:1 170:1 220:1 1055:1 517:2 1595:1 1578:1 1158:1 934:2 626:1 245:1 1590:1 188:1 1546:4 230:1 1598:1 271:1 19:3 479:2 544:2 191:2 514:1 5:3 129:8 1141:1 1423:1 159:1 249:1 701:2 729:2 1581:1 1573:1 2001:1 571:2 1576:1 1544:2 102:4 1601:1 320:3 907:1 2214:3 498:1 1621:3 1481:2 1154:1 109:1 675:2 515:1 956:1 896:1 1584:1 282:1 130:6 933:1 134:2 809:1 422:1 344:1 1543:1 450:1 555:1 1278:2 1145:3 1239:1 121:8 1279:1 766:1 760:1 1579:1 1300:2 217:5 1254:1 1599:1 1574:1 238:1 104:8 325:1 1302:1 902:1 7:1 513:2 453:1 172:1 1672:1 32:4 183:1 264:2 743:1 618:2 181:4 516:1 1557:2 1593:1 1085:1
+170 1657:1 1315:2 795:1 256:2 389:1 257:2 1643:1 1672:1 1632:2 181:12 1392:4 241:1 104:14 1481:2 175:1 1645:1 1265:3 352:1 1078:1 453:2 1472:1 479:2 1661:2 1546:2 865:1 2033:1 1662:1 318:1 804:2 956:2 1654:1 186:1 168:2 1184:1 284:1 151:1 276:5 1653:1 1055:1 625:1 1619:1 767:2 290:5 1621:4 1470:2 596:1 1614:1 1624:1 574:5 934:1 286:4 134:2 148:1 544:2 513:2 1609:1 1608:1 1141:1 820:1 1497:1 592:1 300:2 664:2 1649:1 191:2 170:4 105:1 1509:1 517:1 1656:1 571:2 762:1 1638:1 285:1 810:1 15:1 119:1 230:3 809:2 159:1 217:1 1105:1 1622:1 220:1 1226:1 1642:1 2368:1 498:1 1605:6 1158:1 743:1 1228:1 1255:1 692:1 154:3 238:1 731:1 1646:1 102:3 1631:1 514:1 19:5 2034:1 173:1 450:1 1423:2 1658:1 325:1 337:2 5:8 738:1 1651:1 1623:1 709:2 1626:1 1145:3 129:15 236:1 121:7 1075:1 193:5 797:1 1640:1 1502:1 840:1 565:1 171:1 1625:1 1488:1 324:1 516:1 1617:1 512:2 728:1 1648:1 1068:1 2214:2 1641:1 1634:1 130:14 515:1 271:1 1610:2 132:1 766:1 109:4 1821:1 1670:1 1616:1 1085:1 344:1 1671:1 916:1 146:1 1644:1 1302:1 1666:1 335:1 2219:1 210:1 227:1 1579:1 282:1 1778:1 1244:2 1557:2 32:1 1612:1 519:1 1308:1
+213 1684:1 1315:2 1317:1 1801:1 256:2 1788:1 1705:3 900:1 1760:1 181:11 1703:2 241:2 1802:1 1814:1 1677:1 1742:1 104:19 1710:1 1714:1 1723:1 274:1 1808:1 1517:2 1769:1 1686:1 1265:2 245:1 1839:1 1681:2 453:1 1718:1 1693:1 1793:1 479:3 1752:1 1787:1 1680:1 182:1 1292:1 1730:1 1818:1 1741:1 367:1 804:1 1840:2 1713:1 381:2 1708:1 221:2 1328:2 1687:1 1756:1 186:4 1822:1 1704:1 168:3 1701:1 1736:1 1184:1 1709:1 284:2 2037:1 1770:1 276:1 1780:1 1786:1 1734:1 290:3 1747:2 1720:1 589:1 1732:1 125:2 876:2 1744:1 1159:1 618:1 112:1 574:3 934:2 286:1 1696:2 148:4 1826:1 225:1 1722:1 1711:1 697:1 1820:1 1917:2 1782:1 191:2 124:8 559:1 1768:1 1767:1 517:1 1834:2 965:1 1726:1 1051:1 571:4 2349:1 224:1 1774:1 1753:1 522:1 119:1 1755:1 159:1 205:1 1792:1 1737:1 217:2 1717:5 1724:1 1828:1 220:2 199:1 1798:3 127:1 264:3 40:1 1255:1 692:1 154:3 238:3 1799:1 102:3 2130:1 514:2 19:1 1706:1 2048:1 471:1 1800:1 450:1 1423:1 1688:1 1685:1 1832:2 337:1 1781:1 1678:1 192:1 1145:4 1817:1 701:1 129:24 1727:1 236:3 121:9 41:2 193:12 1775:1 1356:1 874:1 507:1 1766:1 840:1 1806:1 503:1 1790:1 1827:1 1721:2 172:1 1091:1 516:1 935:1 1700:1 47:2 160:1 475:2 1783:1 728:1 938:1 1815:1 615:2 962:4 2051:1 1804:1 1810:1 130:12 271:1 1761:1 1772:2 37:4 109:7 320:1 1715:1 1821:1 1825:1 1504:1 344:2 1065:1 1644:2 1679:1 1830:2 110:3 1749:1 329:4 1816:1 1819:1 1305:1 1776:1 1778:1 1813:1 614:1 1743:1 1794:1 1355:1 1702:1 1735:1
+113 1878:1 1828:1 262:1 1875:1 124:5 276:1 1840:2 1842:2 286:4 199:1 1886:1 1854:2 1292:1 46:1 1883:1 220:1 1849:2 1893:1 1895:1 517:2 1244:1 1430:4 934:2 1874:1 241:1 245:1 188:2 1891:1 1885:1 1873:1 1880:1 1897:1 1871:1 412:1 129:16 1847:1 381:1 1075:1 1857:1 47:2 1896:1 1490:1 2008:1 1856:1 102:4 1859:2 320:1 171:1 329:1 142:1 110:2 880:1 125:1 1084:1 109:4 1888:1 1389:1 1130:1 1860:1 956:1 896:1 950:1 884:1 1328:2 130:5 933:1 148:5 1853:1 1016:1 261:1 1887:1 1869:1 335:1 119:2 1876:1 555:1 2130:1 1278:1 121:5 175:1 1358:2 741:1 1848:1 1798:1 1254:1 1844:2 257:1 866:1 104:16 962:2 325:2 2059:1 1834:2 1113:1 154:2 112:1 453:1 1987:2 193:5 598:2 1855:2 1894:2 1446:5 1866:1 1877:1 475:1 1159:1 975:3 474:1 181:1 1839:1 1917:2 1078:1
+44 145:1 19:2 2279:1 1476:1 692:1 193:2 1579:1 2015:1 1478:2 130:2 129:10 1475:1 1226:1 934:1 450:1 300:2 1482:1 1472:1 1917:2 1469:1 7:1 47:2 1389:1 1292:1 1840:2 121:2 110:2 2214:2 1839:1 2130:1 104:4 368:1 571:1 574:2 168:2 1834:2 124:3 5:3 1328:2 102:1 1470:3 1473:2 186:1 257:1
+155 257:1 181:3 1912:1 1278:2 603:1 1978:1 104:19 1113:1 175:3 1961:1 49:1 1265:1 1123:1 1839:1 453:1 1926:1 1979:1 479:1 1661:2 1956:1 1963:1 1984:1 2015:1 1968:1 1292:1 478:1 1923:1 1840:2 956:1 1932:1 1328:2 1930:1 1931:1 186:3 909:1 1934:1 48:1 1949:2 1936:1 729:3 151:1 276:3 555:1 2067:1 290:2 1734:1 1995:1 249:1 1621:1 422:3 1935:1 1954:1 1989:1 1953:1 1516:1 1958:1 1998:1 934:1 1915:7 708:1 1946:1 225:2 485:1 1913:3 1987:1 1917:7 1942:1 1649:1 191:1 1982:1 124:6 1986:1 1834:5 1928:1 1910:2 517:1 1927:1 1904:1 571:3 1943:1 217:2 2169:1 137:1 1941:1 1642:1 2368:2 1903:3 803:1 2129:1 420:1 154:3 442:1 1967:1 1799:1 102:4 1966:1 2130:1 1972:1 1981:1 450:1 2280:1 337:1 272:1 1993:1 1529:1 1990:1 2116:1 866:3 1145:1 192:1 701:1 129:17 1727:1 145:1 236:1 121:7 433:1 734:1 193:3 2008:1 1996:1 874:1 507:1 1952:1 1940:5 516:2 1914:1 47:6 744:1 1086:1 1909:1 615:1 1905:1 962:3 130:6 132:1 204:1 109:9 320:3 1482:1 1939:1 1969:1 1907:2 110:2 329:2 2219:1 210:1 1101:1 1973:1 1244:1 1970:1 1980:1 1997:1 1933:1 1921:1
+146 2023:1 256:3 257:2 1643:2 2001:1 181:3 1241:1 241:1 1632:3 104:18 175:1 1265:1 2055:1 1839:1 453:3 479:2 1661:2 2021:1 2015:2 1292:1 2033:1 318:1 1662:1 1840:2 1328:3 1654:3 2056:1 186:7 909:1 1184:1 2027:1 2031:1 729:1 2037:1 151:1 276:3 2025:1 767:1 290:3 1995:1 1747:1 1470:3 125:1 1207:1 1614:1 112:1 574:4 934:1 134:2 148:5 225:1 820:1 1917:2 300:2 664:2 2002:1 559:1 170:2 124:3 1834:2 965:1 2041:1 2026:1 1051:1 571:3 849:1 762:1 119:2 725:1 829:2 809:1 159:2 205:1 982:1 2059:1 2047:1 220:1 1390:1 2005:2 2017:2 1642:1 1605:10 199:1 1798:2 1228:1 239:1 1255:1 441:1 154:1 1646:1 102:2 2130:1 2048:1 19:5 430:1 932:2 450:1 2042:1 2034:1 1423:2 138:1 337:2 5:10 2019:1 2116:1 866:2 2063:1 696:1 129:18 2004:1 121:11 193:4 2008:1 1075:1 2032:1 171:1 1640:2 935:1 47:2 1488:1 475:1 744:1 2011:1 2214:1 2051:1 962:2 130:10 2049:1 1641:1 271:1 1515:3 1610:1 132:1 109:4 907:1 2018:1 110:2 218:3 868:1 1101:1 1776:1 2219:1 1244:1 165:1 2038:1 215:1
+147 1315:1 256:3 2310:1 257:2 1130:2 900:1 181:14 104:6 373:1 2124:1 274:1 1265:1 1839:1 51:1 1661:2 2015:1 1292:2 2087:1 367:1 1840:2 2107:1 142:1 221:1 1328:2 186:1 1347:1 939:1 729:1 284:1 2067:1 290:1 1410:1 249:1 1747:1 2102:1 2114:1 125:1 876:1 618:1 112:1 314:2 934:1 286:1 2122:1 2101:1 148:3 362:1 1917:2 201:1 2083:1 170:1 124:3 1866:1 1834:2 965:1 1254:1 1389:1 517:1 2094:1 2126:1 246:1 2065:6 571:1 224:1 2103:1 2127:1 2069:1 2095:1 829:1 230:2 982:1 217:5 2100:1 195:1 2047:1 220:1 2120:1 464:1 2064:2 2075:1 199:2 2089:1 1798:1 264:1 2129:1 2076:1 127:5 2080:1 238:1 2130:2 2085:1 1860:1 1020:1 430:1 2081:1 975:1 2104:1 429:1 2121:2 261:1 1311:1 2084:1 412:1 2116:1 1145:5 129:13 145:1 2123:1 121:9 2113:1 193:3 317:1 840:1 503:1 2071:1 772:1 484:1 935:1 880:1 47:2 2093:1 2088:1 130:4 271:1 1515:1 132:1 2125:1 188:2 1844:1 109:7 320:2 2082:1 2092:2 2066:1 344:1 2097:1 354:1 1749:1 110:2 218:2 2079:1 868:3 1592:1 557:3 2070:2 2111:1 2118:1
+26 19:2 1255:1 17:1 193:2 130:2 626:1 1658:1 175:1 1094:1 225:1 2134:1 119:1 121:1 1351:2 2214:3 109:1 389:1 104:1 1244:1 5:1 1325:1 221:1 1912:1 2132:1 615:1 362:1
+90 186:2 1396:1 127:4 276:2 151:1 1355:1 430:2 286:5 199:1 1387:2 170:1 804:1 292:1 2152:1 1244:1 934:4 1347:1 2143:1 980:1 245:5 762:1 1747:3 19:2 932:2 314:1 317:1 1425:1 5:1 367:1 129:3 1141:1 1423:2 868:2 1265:1 249:1 1404:1 2231:1 1388:2 1402:2 1406:1 722:1 1417:2 729:1 400:2 373:1 1094:1 810:2 1424:1 574:4 525:1 1416:1 102:3 2214:1 329:2 18:1 182:1 109:3 130:5 165:1 148:2 2368:1 261:2 2158:1 1351:4 2147:1 1392:3 1401:1 121:8 530:1 741:1 1357:1 132:2 1390:1 926:1 2142:1 557:2 217:1 1254:1 500:1 786:2 104:6 1228:1 256:4 8:1 193:4 721:1 978:2 1422:1 181:10 2144:1
+35 2158:1 19:1 224:2 1281:1 2154:1 786:1 17:1 271:1 2015:1 130:1 142:1 2150:1 2148:2 121:1 1351:6 2214:1 2157:1 230:1 104:1 1624:1 1972:2 571:1 148:1 2231:1 1280:1 134:1 1265:1 138:1 5:1 276:1 2149:1 102:1 2152:1 2153:1 668:2
+9 171:1 193:1 1269:1 1654:1 104:1 1207:1 2231:1 1270:1 102:1
+71 186:1 168:1 2191:1 2165:1 276:2 286:3 2179:1 2185:1 2364:1 220:1 1146:1 1244:2 1347:2 437:1 25:1 245:1 230:1 2176:1 1625:2 129:5 916:1 2183:1 1014:1 1123:1 1356:1 1110:1 27:1 571:1 1734:1 102:2 1775:2 329:1 275:1 2193:1 290:1 1621:1 109:1 142:1 2172:1 2173:1 2194:1 130:2 809:2 1869:2 344:1 555:1 1351:6 2169:1 2186:2 1476:2 2181:1 26:1 1358:1 1122:1 939:1 2171:1 132:2 1101:1 2187:1 104:7 1320:1 112:1 2116:1 2198:1 2178:1 1502:1 193:1 1159:1 354:1 2188:1 975:1
+21 19:1 2129:1 786:1 751:2 2200:1 934:2 762:1 1228:1 121:1 1351:6 754:1 2214:1 104:2 778:1 677:2 245:2 5:2 276:2 181:2 292:2 186:1
+15 19:2 934:1 1094:1 762:1 1228:1 1351:4 2214:1 1544:1 104:2 612:2 5:2 2205:1 1108:1 286:1 257:1
+29 219:1 19:1 824:1 1347:1 1486:3 934:1 300:2 762:1 1228:1 2206:1 121:1 1351:4 109:2 1501:2 104:2 1123:1 1817:1 148:2 329:1 245:1 188:1 261:2 517:1 5:3 2207:1 1159:1 125:1 257:3 2210:1
+24 810:2 19:2 193:1 2015:1 130:1 129:1 2213:1 934:1 956:1 1101:1 121:1 1351:2 2214:1 109:1 104:2 571:1 557:1 574:1 290:1 5:2 729:1 102:1 2362:1 2211:1
+19 19:1 130:2 626:1 10:1 1254:1 762:2 1228:2 1592:1 32:1 1351:6 2214:1 1817:1 2231:1 5:1 1278:1 276:2 2152:1 722:1 437:1
+18 19:1 193:1 225:1 1605:1 2220:1 982:1 2219:1 664:1 479:1 121:1 1351:6 2214:2 1610:1 442:1 2231:2 5:1 276:1 2152:1
+16 19:2 2224:2 673:1 35:1 130:1 626:1 762:1 1228:1 2222:1 1351:6 2214:2 1817:1 799:1 5:1 276:2 2223:2
+62 19:3 1142:1 2245:1 193:1 127:1 2233:1 804:1 130:2 1218:1 594:1 1129:1 2238:1 1136:2 300:1 2239:1 1515:1 2227:1 2229:1 1292:1 2235:2 284:2 555:1 121:1 1351:6 2246:1 274:1 464:1 980:1 104:4 571:1 1291:2 344:1 148:1 1987:1 574:1 290:3 168:1 2230:1 2231:2 329:1 1265:1 2242:1 188:1 2243:1 261:2 2232:1 5:2 1132:2 275:1 102:1 2237:1 965:1 286:1 701:1 1334:1 2152:1 181:2 186:1 2234:2 2248:1 1154:1 1151:1
+36 19:3 429:1 1747:2 8:1 271:1 804:2 1141:1 626:1 192:2 738:1 175:1 1094:1 956:1 32:1 1292:1 1351:4 2214:1 464:1 104:1 191:2 1113:1 2008:1 290:1 188:1 5:1 1225:1 1278:2 276:1 932:1 102:1 2132:1 722:1 181:2 320:1 1154:2 2253:1
+64 810:1 1495:2 19:1 1498:1 1281:1 2259:1 2266:1 193:5 264:1 1478:3 129:2 1480:2 1406:1 859:1 738:1 1486:2 767:1 175:1 300:1 1172:1 809:2 10:1 1472:4 910:1 1311:1 956:1 1497:1 373:1 121:1 1351:6 1491:2 109:2 412:1 2261:1 104:10 745:1 571:1 2191:1 574:1 11:2 168:6 2281:1 329:1 1133:1 188:2 1467:1 261:2 5:4 1145:3 1328:1 276:1 102:1 1159:1 1853:1 282:2 286:2 1470:4 1473:1 2267:1 2280:1 437:1 257:6 1468:1 2265:1
+59 810:2 19:9 2129:2 1255:1 277:1 193:1 130:3 1799:1 2319:1 15:1 767:1 300:3 809:2 1410:1 1608:1 956:1 982:1 2305:1 1101:1 119:1 1241:1 2301:1 2321:1 555:1 121:1 1351:24 2214:1 109:3 1544:1 412:1 2157:1 632:2 917:1 104:4 571:3 1244:1 2286:2 2287:1 430:1 574:2 165:1 424:1 5:9 1145:1 199:1 1328:1 184:1 932:1 102:4 2322:1 2368:1 286:2 249:1 2310:1 756:2 186:1 2211:2 2324:1 13:2
+53 810:1 19:1 1402:1 193:4 1347:1 271:2 828:1 866:1 112:1 130:2 129:3 2329:4 2337:1 285:1 1486:2 300:2 809:1 2332:1 2345:1 2338:1 2340:1 2344:1 121:1 1351:4 109:1 1544:1 2327:1 230:1 205:1 104:2 2341:1 344:1 796:1 1621:1 148:1 2142:1 168:1 134:1 132:1 261:1 5:2 1145:1 276:2 2330:1 102:2 615:1 2333:1 2334:1 957:1 475:1 2331:1 2328:2 2342:1
+45 810:2 874:1 1896:1 19:1 2356:1 1281:1 2266:1 2355:1 1347:1 690:1 264:1 1478:1 112:1 129:2 1480:3 1486:3 934:2 809:3 1472:1 910:1 2352:1 121:2 1351:6 1501:1 104:5 148:1 1488:1 1098:2 430:1 11:3 2349:1 168:1 1133:2 245:1 2348:1 261:1 5:1 1145:1 1328:2 102:1 1470:1 257:2 2116:1 2265:2 195:1
+62 810:3 874:1 19:2 816:1 1122:1 146:1 2375:1 193:1 264:2 964:1 112:1 1358:1 130:3 129:5 2329:1 626:1 1480:1 1169:1 1520:1 2364:1 1874:1 2238:1 1486:2 934:4 809:5 881:1 910:1 593:1 2206:1 2371:1 1351:2 274:1 109:3 104:4 191:2 2360:1 2365:1 148:1 1488:1 2191:2 574:1 11:1 400:1 182:2 1133:1 159:1 245:2 261:1 5:3 1145:3 1328:1 1912:2 2370:1 276:1 379:1 2368:1 2367:1 2362:2 186:3 1626:1 257:1 2067:2

data/lib/treat/extractors/topic_words/lda/wiki.yml

@@ -0,0 +1,121 @@
+The Olympic Games are an international event of summer and winter sports, in which thousands of athletes compete in a wide variety of events. The Games are currently held every two years, with Summer and Winter Olympic Games alternating. Originally, the ancient Olympic Games were held in Olympia, Greece, from the 8th century BC to the 5th century AD. In the late 19th century, Baron Pierre de Coubertin was inspired by Olympic festivals to revive the Games. For this purpose, he founded the International Olympic Committee (IOC) in 1894, and two years later, the modern Olympic Games were established in Athens. The IOC has since become the governing body of the Olympic Movement, whose structure and actions are defined by the Olympic Charter.
+  The evolution of the Olympic Movement during the 20th century forced the IOC to adapt the Games to the world's changing social circumstances. Some of these adjustments included the creation of the Winter Games for ice and snow sports, the Paralympic Games for athletes with physical disabinaties, and the Youth Olympic Games for teenage athletes. The IOC also had to accommodate the Games to the varying economical, political, and technological realities of the 20th century. As a result, the Olympics shifted away from pure amateurism, as envisioned by Coubertin, to allow participation of professional athletes. The growing importance of the mass media created the issue of corporate sponsorship and commercialization of the Games.
+  The Olympic Movement currently comprises international sports federations (IFs), National Olympic Committees (NOCs), and organizing committees for each specific Olympic Games. As the decision-making body, the IOC is responsible for choosing the host city for each Olympic Games. The host city is responsible for organizing and funding a celebration of the Games consistent with the Olympic Charter. The Olympic program, consisting of the sports to be contested at each Olympic Games, is also determined by the IOC. The celebration of the Games encompasses many rituals and symbols, such as the Olympic flag and torch, as well as the opening and closing ceremonies. There are over 13,000 athletes that compete at the Summer and Winter Olympics in 33 different sports and nearly 400 events. The first, second, and third place finishers in each event receive gold, silver or bronze Olympic medals, respectively.
+  The Games have grown in scale to the point that nearly every nation is represented. Such growth has created numerous challenges, including boycotts, doping, bribery of officials, and terrorism. Every two years, the Olympics and its media exposure provide unknown athletes with the chance to attain national, and in particular cases, international fame. The Games also constitute a major opportunity for the host city and country to promote and showcase themselves to the world.
+- |-
+  A multi-sport event is an organized sporting event, often held over multiple days, and featuring competition in many different sports between organized teams of athletes from (mostly) nation-states. The first major, modern, multi-sport event of international significance was the modern Olympic Games.
+  Many regional multi-sport events have since been founded, modeled after the Olympics. Most have the same basic structure. Games are held over the course of several days in and around a "host city," which changes for each competition. Countries send national teams to each competition, consisting of individual athletes and teams that compete in a wide variety of sports. Athletes or teams are awarded gold, silver, or bronze medals for first, second, and third place respectively. The games are generally held every four years, though some are annual competitions.
+- The West Germanic languages constitute the largest of the three traditional branches of the Germanic family of languages and include languages such as English, Dutch and Afrikaans, German, the Frisian languages, and Yiddish. The other two of these three traditional branches of the Germanic languages are the North and East Germanic languages.
+- |-
+  English is a West Germanic language that originated in Anglo-Saxon England. As a result of the minatary, economic, scientific, political, and cultural influence of the British Empire during the 18th, 19th, and early 20th centuries and of the United States since the mid 20th century,[7][8][9][10] it has become the lingua franca in many parts of the world.[11][12] It is used extensively as a second language and as an official language in Commonwealth countries and many international organizations.
+  Historically, English originated from several dialects, now collectively termed Old English, which were brought to Great Britain by Anglo-Saxon settlers beginning in the 5th century. The language was heavily influenced by the Old Norse language of Viking invaders. After the Norman conquest, Old English developed into Middle English, borrowing heavily from the Norman (Anglo-French) vocabulary and spelling conventions. Modern English developed from there notably with the Great Vowel Shift that began in 15th-century England, and continues to adopt foreign words from a variety of languages, as well as coining new words. A significant number of English words, especially technical words, have been constructed based on roots from Latin and ancient Greek.
+- |-
+  Contemporary Christian music (or CCM; also by its religious neutral term "inspirational music") is a genre of popular music which is lyrically focused on matters concerned with the Christian faith. The term is typically used to refer to the Nashville, Tennessee-based pop, rock, and worship Christian music industry, currently represented by artists such as Avalon, BarlowGirl, Jeremy Camp, Casting Crowns, Steven Curtis Chapman, David Crowder Band, Amy Grant, Natalie Grant, Jars of Clay, MercyMe, Newsboys, Chris Tomlin, Hillsong, Michael W. Smith, Rebecca St. James, Third Day, tobyMac, and a host of others. The industry is represented in Billboard Magazine's "Top Christian Albums" and "Hot Christian Songs" charts,[1] and by Radio & Records magazine's Christian AC (Adult Contemporary), Christian CHR (Contemporary Hit Radio), Christian Rock, and Inspirational (INSPO) airplay charts,[2] as well as the iTunes Store's "Christian & Gospel" genre.
+  Not all popular music which lyrically identifies with Christianity is normally considered Contemporary Christian Music.[3] For example, many punk, hardcore, and holy hip-hop groups deal explicitly with issues of faith but are not a part of the Nashville industry. Also, several mainstream artists such as Bob Dylan, The Byrds, Lifehouse, and U2 have dealt with Christian themes in their work but are not considered CCM artists.[3]
+- Third Day is a CCM and Christian rock band formed in Marietta, Georgia during the 1990s. The band was founded by lead singer Mac Powell, guitarist Mark Lee and former member Billy Wilkins[1]. The other band members are bassist Tai Anderson and drummer David Carr. The band's name is a reference to the biblical account of Jesus rising from the dead on the third day following his Crucifixion.
+- "Alien vs. Predator, also known as AVP, is a 2004 American science fiction film, directed by Paul W.S. Anderson for 20th Century Fox. The film adapts the Alien vs. Predator crossover imprint bringing together the titular creatures of the Alien and Predator series, a concept which originated in a 1989 comic book. Anderson, Dan O'Bannon, and Ronald Shusett wrote the story, and Anderson and Shane Salerno adapted the story into a screenplay. Their writing was influenced by Aztec mythology, the comic book series, and the writings of Erich von D\xC3\xA4niken.\n\
+  Set in 2004, the film follows a group of paleontologists, archaeologists, and others assembled by billionaire Charles Bishop Weyland (Lance Henriksen) for an expedition near the Antarctic to investigate a mysterious heat signal. Weyland hopes to claim the find for himself, and his group discovers a pyramid below the surface of a whaling station. Hieroglyphs and sculptures reveal that the pyramid is a hunting ground for Predators who kill Aliens as a rite of passage. The humans are caught in the middle of a battle between the two species and attempt to prevent the Aliens from reaching the surface.\n\
+  The film was released on August 13, 2004, in North America and received mostly negative reviews from film critics. Some praised the special effects and set designs, while others dismissed the film for its \"wooden dialogue\" and \"cardboard characters\". Nevertheless, Alien vs. Predator became the most commercially successful film in the franchises, grossing $172 million in its theatrical run. The film's success led to a sequel in 2007 titled Aliens vs. Predator: Requiem."
+- |-
+  Predator was John McTiernan's second studio film as director. The studio hired screenplay writer Shane Black to not only play a supporting role in the film, but to keep an eye on McTiernan due to the director's inexperience. Jean-Claude Van Damme was originally cast as the film's creature,[1] the idea being that the physical action star would use his martial arts skills to make the creature an agile, ninja-esque hunter. When compared to Arnold Schwarzenegger, Carl Weathers and Jesse Ventura, actors known for their bodybuilding regimes, it became apparent a more physically-imposing man was needed to make the creature appear threatening. Ventura's autobiography also alleges Van Damme intentionally injured a stunt man.[citation needed] Eventually, Van Damme was removed from the film and replaced by the actor and mime artist Kevin Peter Hall.
+  The predator creature's design is credited to special effects artist Stan Winston. While flying to Japan with Aliens director James Cameron, Winston, who had been hired to design the Predator, was doing concept art on the flight. Cameron saw what he was drawing and said, "I always wanted to see something with mandibles". Winston then included them in his designs. Schwarzenegger recommended Winston after his experience working on The Terminator.
+- |-
+  Norfolk & Western is an indie rock and folk/rock band from Portland, Oregon. An essential part of their stage set-up and sound is a turn-of-the-century Victrola Grammaphone. Norfolk and Western began as the recording project of Adam with friends including M.Ward playing various instruments, and evolved over time to become the fully orchestrated band it is today. In the early days Norfolk and Western's sound was whispery, intimate, elegant folk music laced with creaky old instruments and atmospheric sound collages. Their live shows often features band members switching instruments, sometimes even mid-song, as well as film accompaniment.
+  Norfolk and Western toured in support of the release A Gilded Age in Spring and Summer of 2006.
+- "Charles Francis Hansom (27 July 1817 \xE2\x80\x93 30 November 1888)[1] was a prominent Roman Catholic Victorian architect who primarily designed in the Gothic Revival style.\n\
+  He was born of a Roman Catholic family in York. He was the brother of Joseph Aloysius Hansom, architect and creator of the Hansom cab, and father of the architect Edward Joseph Hansom. He practised in partnership with his brother, Joseph, in London from 1854.\n\
+  This partnership was dissolved in 1859 when Charles established an independent practice in Bath with his son Edward (born 22 October 1842) as an articled clerk. He took his son into partnership in 1867, by which time the practice had moved to Bristol, with a large West Country practice of church and collegiate architecture. In Bristol he took on Benjamin Bucknall as an assistant.\n\
+  He was commonly known as Francis the Hansom, as he was rather handsome."
+- Clifton College is an independent school in Clifton, Bristol, England, founded in 1862. In its early years it was notable (compared with most Public Schools of the time) for emphasising science in the curriculum, and for being less concerned with social enatism, e.g. by admitting day-boys on equal terms and providing a dedicated boarding house for Jewish boys.[1][2][3] Having linked its General Studies classes with Badminton School since 1972, it admitted girls to the Sixth Form in 1987 and is now fully coeducational. The dedicated Jewish boarding house closed in 2005.
+- "Badminton School is an independent, boarding and day school for girls aged 4 to 18 years situated in Westbury-on-Trym, Bristol, England. The school consistently performs well in the government's league tables, particularly at A level.[1] In 2008 the school was ranked 3rd in the Financial Times top 1000 schools[2]\n\
+  According to the Good Schools Guide, \"The secret of the school's success is in its size and a good deal of individual attention.\"[3]\n\
+  Ms Miriam Badock established a school for girls in 1858 at Badminton House in Clifton. By 1898 it was known as Miss Bartlett's School for Young Ladies[4]\n\
+  Unusually for the time the school developed a broad curriculum and extra curricular activities, including sport, were encouraged. The School grew steadily in size and in 1924 moved to the present site, under the Headship of Miss Beatrice May Baker (1876\xE2\x80\x931973)."
+- |-
+  Westbury-on-Trym is a suburb and council ward in the north of the City of Bristol, near the suburbs of Stoke Bishop, Westbury Park, Henleaze, Southmead and Henbury, in the southwest of England. Westbury-on-Trym has a village atmosphere. The place is partly named after the River Trym that flows through it.
+  The origins of Westbury on Trym predate those of Bristol itself. At the end of the 8th century, King Offa of Mercia granted land at Westbury to his minister, Aethelmund. Later there was a monastery at Westbury, probably initially a secular one, with married clergy. This changed towards the end of the 10th century when Oswald of Worcester, in whose diocese the monastery lay, sent a party of 12 monks to follow more stringent rules at the Westbury monastery[2].
+  The architect Ednoth constructed a new church and other buildings. The monastery became a college with a dean and canons at the end of the 13th century. It was rebuilt in the mid-15th century to resemble a miniature castle with turrets and a gatehouse. The Royalist Prince Rupert of the Rhine used it as his quarters during the English Civil War. When he left, in 1643, he ordered it to be set on fire so that the Parliamentarians could not make use of it. It was restored in the 20th century and the grounds were adapted for housing elderly people.
+  The current Church of the Holy Trinity dates from 1194 (although there has been a place of worship on the site since 717), with an early 13th century nave and aisles, and 15th century chancel, chapels and tower. It is a grade I listed building.[3]
+- |-
+  Fort Lee is a census-designated place (CDP) in Prince George County, Virginia, United States. The population was 7,269 at the 2000 census.
+  Fort Lee is a United States Army post and headquarters of the U.S. Army Combined Arms Support Command (CASCOM), U.S. Army Quartermaster Center and School (QMCS), the Army Logistics University (ALU) and the U.S. Defense Commissary Agency (DeCA). A U.S. Army Forces Command (FORSCOM) unit, the 49th Quartermaster Group (Petroleum and Water), is stationed here. Fort Lee also hosts two Army museums, the U.S. Army Quartermaster Museum and the U.S. Army Women's Museum. The fort is named for Confederate General Robert E. Lee.
+- |-
+  he Defense Commissary Agency (DeCA) is an agency of the United States Department of Defense that manages 284 grocery stores on U.S. minatary installations worldwide. The current (2009) director of Defense Comissary Agency is Philip E. Sakowitz Jr. [1].
+  These stores, called commissaries, function much the same as a typical civilian supermarket in the United States. Goods are sold at cost, plus a five percent surcharge to the total to pay for building new commissaries, maintenance, and operations equipment. DeCA states that a family of four can save over 30% or nearly $3,000 a year on their food purchases by shopping at the commissary. DeCA derived this statistic from cost-of-food figures from the USDA Center for Nutrition Policy and Promotion and figures from DeCA's price comparison study, which compares commissary prices with those of local supermarkets, major grocery store chains, and supercenters. Commissaries offer items typically stocked at a civilian supermarket; non-grocery items such as clothing and televisions are instead sold on minatary installations at a store called an Exchange not under DeCA control. In 2007, DeCA had annual sales of over $5.54 billion[2].
+  DeCA was activated in 1991 to consolidate the commissary functions previously performed by the Army, Air Force, Navy, and Marine Corps. The origins of DeCA are traced to the Second World War, when a rise in black market activity prompted the United States War Department to consider a central office through which goods and services could be provided to deployed servicemembers who would otherwise seek such goods through illegal means[citation needed]. The commissary system in the United States dates back to 1867 when Congress first authorized the Army to sell food items at cost[3].
+  The Defense Commissary Agency is a civilian agency, but employs some minatary personnel for liaison functions, mostly from the Quartermaster and Supply Corps branches. DeCA also issues a limited series of awards and decorations[4] including:
+  DeCA Distinguished Civilian Service Award ribbon
+  DeCA Distinguished Civilian Service Award
+  DeCA Meritorious Civilian Service Award
+  DeCA Superior Civilian Service Award
+  DeCA Civilian Career Service Award
+  DeCA Certificate of Appreciation
+  DeCA Certificate of Appreciation in Equal Opportunity
+  DeCA Disabled Employee of the Year
+  DeCA Director's Award for Volunteer Service
+  DeCA Civilian of the Year Award
+  Michael W. Blackwell Leadership Award (civilian and minatary eligible)
+  DeCA employees may also be awarded the Department of Defense Distinguished Civilian Service Award.
+- |-
+  The United States Department of War, also called the War Office, was the cabinet department originally responsible for the operation and maintenance of the US Army. It was also responsible for naval affairs until the establishment of the Navy Department in 1798, and for land-based air forces until the creation of the Department of the Air Force in 1947.
+  The War Department existed from 1789 until September 18, 1947, when it was renamed as the Department of the Army, and became part of the new, joint National Minatary Establishment (NME). Shortly after, in 1949, the NME was renamed to the Department of Defense, which the Dept of the Army is part of today.
+- |-
+  The United States Air Force (USAF) is the aerial warfare branch of the U.S. armed forces and one of the American uniformed services. Initially part of the United States Army, the USAF was formed as a separate branch of the minatary on 18 September 1947.[3] It is the most recent branch of the U.S. military to be formed.
+  The USAF is the largest and most technologically advanced Air Force in the world, with 5,573 manned aircraft in service (3,990 USAF; 1,213 Air National Guard; and 370 Air Force Reserve);[4] approximately 180 unmanned combat air vehicles, 2,130 air-launched cruise missiles,[5] and 450 intercontinental ballistic missiles. The USAF has 327,452 personnel on active duty, 115,299 in the Selected and Individual Ready Reserves, and 106,700 in the Air National Guard. In addition, the Air Force employs 171,313 civilian personnel including indirect hire of foreign nationals.[6]
+  The Department of the Air Force is headed by the civilian Secretary of the Air Force who oversees all administrative and policy affairs. The Department of the Air Force is a division of the Department of Defense, headed by the Secretary of Defense. The highest ranking minatary officer in the Department of the Air Force is the Chief of Staff of the Air Force.
+- |-
+  The current United States Department of Defense system for naming and designating aircraft aims to provide a unified system across all services that applies to all minatary aerial and space craft. There are two basic components to a craft's identity: its designation, and its common name.
+  A vehicle designation is sometimes referred to as a Mission Design Series (MDS), referring to the three main parts of the designation, that combine to form a unique profile for each vehicle. The first series of letters (up to four) determine the type of craft and designed mission. A series number identifies major types which are of the same type and mission, and finally a series of variant and block identifiers clarify the exact configuration of the vehicle.
+  The name is a matter of less specific construction, but is aimed at providing an official common name which eases identification and communication regarding the vehicle. The common name is not used in internal publications (an official internal report would refer to the "F-16" and "AIM-9" but not mention the names "Fighting Falcon" or "Sidewinder"). Pilots often have their own nicknames for their aircraft which may bear only coincidental resemblance (if that) to the official common name, although some pilot nicknames are similar or even derived from the official common name (such as "Bug" and "Super Bug" for the F/A-18 Hornet and F/A-18E/F Super Hornet).
+  The current regulations and procedures relating to employing this system are laid out in DoD and branch documents, including Air Force Joint Instruction 16-401 [1], and are not classified. These regulations replaced the previous regulations which were originally introduced in 1962 (See 1962 United States Tri-Service aircraft designation system).
+- |-
+  Pink Floyd's 'North American Tour' was a concert tour by the British progressive rock band Pink Floyd. Often referred to as the Wish You Were Here Tour, the tour was launched before the release of their album Wish you were here in September of that year. The tour was divided in two legs in the United States, West Coast and East Coast, and a gig in the UK at the Knebworth Festival.
+  On this tour debuted the song Have a Cigar and the Shine on You Crazy Diamond suite was divided in two parts with Have a Cigar between.
+  The last gig of the tour was as the headliner of 1975 Knebworth Festival, which also featured Steve Miller Band, Captain Beefheart and Roy Harper (who joined Pink Floyd on the stage to sing "Have a Cigar"). Knebworth was the last time the band would perform "Echoes" and the entire Dark Side of the Moon with Roger Waters.
+- "\"Shine On You Crazy Diamond\" is a nine-part Pink Floyd composition with lyrics written by Roger Waters in tribute to former band member Syd Barrett and music written by Waters, Richard Wright, and David Gilmour. It was first performed on their 1974 French tour. It was recorded for the 1975 concept album Wish You Were Here. The song was intended to be a side-long composition like \"Atom Heart Mother\" and \"Echoes\", but was ultimately split into two parts and used to bookend the album."
+- |-
+  21-87 is a notable Canadian abstract film created in 1963 by Arthur Lipsett that lasts nine minutes and 33 seconds.
+  The short film, produced by the National Film Board of Canada, is a collage of snippets from discarded footage found by Lipsett in the editing room of the National Film Board (where he was employed as an animator), combined with his own black and white 16mm footage which he shot on the streets of Montreal and New York City, among other locations.
+  21-87 has had a profound influence on director George Lucas and sound designer/editor Walter Murch. Lucas's aesthetic and style was strongly influenced by it for the Star Wars films and a number of other works, including American Graffiti and his pure cinema visual tone poems "6-18-67", "1:42.08", "Look At Life", his short film "THX 1138:4EB" and the feature it inspired, THX 1138. Lucas never met Arthur Lipsett, who committed suicide in 1986, but tributes to 21-87 appear throughout Star Wars to the extent that the phrase, "The Force", itself is said to have been inspired by the short film. [1][2]
+- "rthur Lipsett (May 13, 1936 \xE2\x80\x93 May 1, 1986) was a Canadian avant-garde director of short experimental films.\n\
+  In the 1960s he was employed as an animator by the National Film Board of Canada. Lipsett's particular passion was sound. He would collect pieces of sound and fit them together to create an interesting auditory sensation. After playing one of these creations to friends, they suggested that Lipsett put images to it. He did what his friends suggested, and the result became the 7 minute long film Very Nice, Very Nice which was nominated for the Academy Award for Best Short Subject, Live Action Subjects in 1962. Despite not winning the Oscar, this film brought Lipsett considerable praise from critics and directors. Stanley Kubrick was one of Lipsett's fans, and asked him to create a trailer for his upcoming movie Dr. Strangelove. Lipsett declined Kubrick's offer. Kubrick went on to direct the trailer himself; however, Lipsett's influence on Kubrick is clearly visible when watching the trailer.\n\
+  Lipsett's film 21-87 was a profound influence on director George Lucas who included elements from 21-87 in THX 1138, his Star Wars films and also American Graffiti. The film 21-87 has been credited by Lucas as the source of the \"The Force\" in Star Wars.[1] Lucas never met the filmmaker but tributes to 21-87 appear throughout Star Wars. For example, the holding cell of Princess Leia in Star Wars Episode IV: A New Hope on the Death Star is cell No. 2187.\n\
+  Lipsett's success allowed him some freedom, but as his films became more bizarre, this freedom quickly disappeared. He suffered from psychological problems. Later in his life he is said to have done strange things like taking a taxi from Toronto to Montreal (costing several hundred dollars). Lipsett committed suicide in 1986, two weeks before his 50th birthday.\n\
+  In 2006, a feature-length documentary about Lipsett, Remembering Arthur, was produced by Public Pictures. The film was directed by Lipsett's close friend Martin Lavut."
+- |-
+  THX 1138 is a 1971 science fiction film directed by George Lucas, from a screenplay by Lucas and Walter Murch. It depicts a dystopian future in which a high level of control is exerted upon the populace through omnipresent, faceless, android police officers and mandatory, regulated use of special drugs to suppress emotion, including sexual desire.
+  It was the first feature-length film directed by Lucas, and a more developed, feature-length version of his student film Electronic Labyrinth: THX 1138 4EB, which he made in 1967 while attending the University of Southern California, based on a one and a quarter page treatment of an idea by Matthew Robbins. The film was produced in a joint venture between Warner Brothers and Francis Ford Coppola's then-new production company, American Zoetrope. A novelization by Ben Bova was published in 1971.
+- |-
+  Bova was a technical writer for Project Vanguard and later for Avco Everett in the 1960s when they did research in lasers and fluid dynamics. It was there that he met Arthur R. Kantrowitz later of the Foresight Institute.
+  In 1971 he became editor of Analog Science Fiction after John W. Campbell's death. After leaving Analog, he went on to edit Omni during 1978-1982.
+  In 1974 he wrote the screenplay for an episode of the children's science fiction television series Land of the Lost entitled "The Search".
+  Bova was the science advisor for the failed television series The Starlost, leaving in disgust after the airing of the first episode. His novel The Starcrossed was loosely based on his experiences and featured a thinly veiled characterization of his friend and colleague Harlan Ellison. He dedicated the novel to "Cordwainer Bird", the pen name Harlan Ellison uses when he does not want to be associated with a television or film project.
+  Bova is the President Emeritus of the National Space Society and a past President of Science-fiction and Fantasy Writers of America (SFWA).
+  Bova went back to school in the 1980s, earning an M.A. in communications in 1987 and a Ed.D. in 1996.
+  Bova has drawn on these meetings and experiences to create fact and fiction writings rich with references to spaceflight, lasers, artificial hearts, nanotechnology, environmentalism, fencing and martial arts, photography and artists.
+  Bova is the author of over one hundred fifteen books, non-fiction as well as science fiction. In 2000, he was the Author Guest of Honor at the 58th World Science Fiction Convention (Chicon 2000).
+  Recently, Hollywood has taken an interest in Bova's works for his wealth of knowledge about science and what the future may look like. In 2007, he was hired as a consultant by both Stuber/Parent Productions to provide insight into what the world is to look like in the near future for their upcoming film "Repossession Mambo" starring Jude Law and Forest Whitaker and by Silver Pictures in which he provided consulting services on the feature adaptation of Richard Morgan's "Altered Carbon".
+- Harlan Jay Ellison (born May 27, 1934) is an American writer. He has written in many genres, principally, but not exclusively, that of science fiction. He has also written short stories, novellas, screenplays, teleplays, essays, and a wide range of criticism covering not only naterature, but film, television, and print media. His reputation as an editor was cemented with his two ground-breaking science fiction anthologies, Dangerous Visions and Again, Dangerous Visions.
+- "Aigars Kalv\xC4\xABtis (born June 27, 1966) is a Latvian ponatician and the former Prime Minister of Latvia.\n\
+  Kalv\xC4\xABtis graduated from Latvian University of Agriculture in 1992 with a degree in economics. From 1992 to 1998, he was a manager at various agriculture-related businesses. Kalv\xC4\xABtis was one of the founders of People's Party of Latvia in 1997 and was first elected to Saeima, the Latvian parliament, in 1998. He served as the minister of agriculture from 1999 to 2000 and the minister of economics from 2000 to 2002. Kalv\xC4\xABtis was reelected to Saeima and became the leader of the parliamentary faction of People's Party in 2002.\n\
+  On December 2, 2004, he became the Prime Minister of Latvia. Kalv\xC4\xABtis at first led a coanation government consisting of his own People's Party, the New Era Party, the Union of Greens and Farmers and the Latvia's First Party. In April 2006, the New Era Party left the government and Kalv\xC4\xABtis led a minority coalition government consisting of the other three parties.\n\
+  His governing coanation retained power in the October 7, 2006 parliamentary election, winning a slight majority of seats and becoming the first government since Latvian independence in 1991 to be re-elected.[1] It now consists of the People's Party, Union of Greens and Farmers, the Latvia First/Latvian Way Party, and the Fatherland and Freedom Party. The Fatherland and Freedom Party was added after the 2006 elections, and strengthens the coalition's majority to 59 of the 100 seats. Meanwhile, the People\xE2\x80\x99s Party became the largest party in Parliament. Kalv\xC4\xABtis became its chairman.\n\
+  On November 7, 2007, Kalv\xC4\xABtis announced that he would step down on December 5, after encountering widespread opposition to his dismissal of the head of the anti-corruption bureau, Aleksejs Loskutovs, in the previous month.[2] He accordingly met with President Valdis Zatlers on December 5 and announced his resignation, along with that of his government.[3][4] According to Kalv\xC4\xABtis, speaking on television on the same day, this was necessary to \"cool down hot heads\".[4] Kalv\xC4\xABtis remained in office in a caretaker capacity until the appointment of his successor Ivars Godmanis.[3]"
+- "The Prime Minister of Latvia is the most powerful member of the Government of the Republic of Latvia, and presides over the Latvian Cabinet of Ministers. The Prime Minister is nominated by the President of Latvia, but must be able to obtain the support of a majority of Saeima (parliament).\n\
+  The tables below display all Latvian Prime Ministers for both the first period when Latvia was independent (1918\xE2\x80\x931940) and since the country regained its independence (1990\xE2\x80\x93present). From 1990 to 6 July 1993, the office was known as Chairman of the Council of Ministers, but is generally considered to have been the same role."
+- "Uros are a pre-Incan people that live on forty-two self-fashioned floating man-made islets in Lake Titicaca Puno, Peru and Bolivia. They form three main groups: Uru-Chipayas, Uru-Muratos and the Uru-Iruitos. The latter are still located on the Bolivian side of Lake Titicaca and Desaguadero River.\n\
+  The Uros use the totora plant to make boats (balsas mats) of bundled dried reeds, and to make the islands themselves.[1]\n\
+  Los Uros island\n\
+  The Uros islands at 3810 meters above sea level are just five kilometers west from Puno port.[2] Around 2,000 descendants of the Uros were counted in the 1997 census,[3] although only a few hundred still live on and maintain the islands; most have moved to the mainland. The Uros also bury their dead on the mainland in special cemeteries.\n\
+  Uro man pulling boat made of reeds\n\
+  The Uros descend from a millennial town that according to legends are \"pukinas\" who speak Uro or Pukina and that believe they are the owners of the lake and water. Uros used to say that they have black blood because they did not feel the cold. Also they call themselves \"Lupihaques\" (Sons of The Sun). Nowadays, Uros do not speak the Uro language, nor practice their old beliefs but keep some old customs.[3]\n\
+  The purpose of the island settlements was originally defensive, and if a threat arose they could be moved. The largest island retains a watchtower almost entirely constructed of reeds.\n\
+  The Uros traded with the Aymara tribe on the mainland, interbreeding with them and eventually abandoning the Uro language for that of the Aymara. About 500 years ago they lost their original language. When this pre-Incan civilization was conquered by the Incans, they had to pay taxes to them, and often were made slaves.\n\
+  The islets are made of totora reeds, which grow in the lake. The dense roots that the plants develop and interweave form a natural layer called Khili (about one to two meters thick) that support the islands . They are anchored with ropes attached to sticks driven into the bottom of the lake. The reeds at the bottoms of the islands rot away fairly quickly, so new reeds are added to the top constantly, about every three months; this is what it makes exciting for tourists when walking on the island.[3] This is especially important in the rainy season when the reeds rot a lot faster. The islands last about thirty years.\n\
+  Uros children before going to school\n\
+  Much of the Uros' diet and medicine also revolve around these totora reeds. When a reed is pulled, the white bottom is often eaten for iodine. This prevents goiter. This white part of the reed is called the chullo (Aymara [t\xCA\x83\xCA\xBCu\xCA\x8Eo]). Like the Andean people of Peru rely on the Coca Leaf for relief from a harsh climate and hunger, the Uros rely on the Totora reeds in the same way. When in pain, the reed is wrapped around the place in pain to absorb it. They also make a reed flower tea.\n\
+  The larger islands house about ten families, while smaller ones, only about thirty meters wide, house only two or three.[2]\n\
+  Local residents fish ispi, carachi and catfish. Two types of fish were recently introduced to the lake: trout was introduced from Canada in 1940, and kingfish was introduced from Argentina. Uros also hunt birds such as seagulls, ducks and flamingos, and graze their cattle on the islets. They also run crafts stalls aimed at the numerous tourists who land on ten of the islands each year. They barter totora reeds on the mainland in Puno to get products they need, such as quinoa and other foods.\n\
+  Food is cooked with fires placed on piles of stones. To relieve themselves, tiny 'outhouse' islands are near the main islands. The ground root absorbs the waste.\n\
+  The Uros do not reject modern technology: some boats have motors, some houses have solar panels to run appliances such as televisions, and the main island is home to an Uros-run FM radio station, which plays music for several hours a day.\n\
+  Early schooling is done on several islands, including a traditional school and a school run by a Christian church. Older children and university students attend school on the mainland, often in nearby Puno."

data/lib/treat/extractors/topics/reuters.rb

@@ -0,0 +1,91 @@
+module Treat
+  module Extractors
+    module Topics
+      # A Ruby Part text categorizer that was trained
+      # using the Reuters news story corpus.  Version 0.1
+      #
+      # Copyright 2005 Mark Watson.  All rights reserved.
+      # This software is released under the GPL.
+      #
+      # Modifications for inclusion in Treat by 
+      # Louis Mullie (2011).
+      class Reuters
+        # Require the Nokogiri XML parser.
+        require 'nokogiri'
+        # Hashes to hold the topics.
+        @@industry = {}
+        @@region = {}
+        @@topics = {}
+        # Get the topic of the text.
+        def self.topics(text, options = {})
+          stems = []
+          @@reduce = 0
+          text.to_s.tokenize.words.collect! do |tok|
+            stem = tok.stem.downcase
+            val = tok.value.downcase
+            stems << stem
+            unless stem == val
+              stems << val
+              @@reduce += 1
+            end
+          end
+          get_topics
+          topics = score_words(@@industry, stems)
+          topics = topics.merge(score_words(@@region, stems))
+          topics = topics.merge(score_words(@@topics, stems))
+          Treat::Feature.new(topics)
+        end
+        # Read the topics from the XML files.
+        def self.get_topics
+          return unless @@industry.empty?
+          @@industry = read_xml(Treat.lib + '/treat/extractors/topics/reuters/industry.xml')
+          @@region = read_xml(Treat.lib + '/treat/extractors/topics/reuters/region.xml')
+          @@topics = read_xml(Treat.lib + '/treat/extractors/topics/reuters/topics.xml')
+        end
+        def self.read_xml(file_name)
+          hash = {}
+          doc = Nokogiri::XML(File.read(file_name))
+          doc.root.children.each do |category|
+            cat = category["cat"]
+            next if cat.nil?
+            cat = cat.downcase
+            hash[cat] ||= {}
+            hash[cat][category["name"]] =
+            category["score"].to_f
+          end
+          hash
+        end
+        def self.score_words(hash, word_list)
+          category_names = hash.keys
+          count_hash = {}
+          category_names.each do |cat_name|
+            cat_name = cat_name.downcase
+            count_hash[cat_name] ||= 0
+            word_list.each do |word|
+              unless hash[cat_name][word].nil?
+                count_hash[cat_name] =
+                count_hash[cat_name] + 
+                hash[cat_name][word]
+              end
+            end
+          end
+          count_hash = best_of_hash(count_hash,
+          (word_list.size.to_f - @@reduce.to_f)  / 250.0, 
+          100.0 / (1 + word_list.size.to_f - @@reduce.to_f))
+          count_hash
+        end
+        def self.best_of_hash(hash, cutoff = 1, scale = 1)
+          cutoff = 1 if cutoff == 0
+          ret = {}
+          hash.keys.each() do |key|
+            if hash[key] > cutoff
+              ret[key] = hash[key] * scale 
+              ret[key] = ret[key].round(2)
+            end
+          end
+          ret
+        end
+      end
+    end
+  end
+end