fix readme

data/nell.test.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa48d27aba4c3b82c2369313fc7a0436103fcb27115efc4159e8b50d988a6fa3
-size 510425
+oid sha256:b60c8ecdf2b65507920b8f11a174d0bc06edc515b2362c4be65aab13fbdb2944
+size 309267

data/nell.train.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc4c73fdae5db374da7855e421176dc09462cc72366040d39ce42d7ec333460d
-size 1997185
+oid sha256:e63db71dd103c5cb9bb96650794dcf0d198e28d737ee8f07bd73e6961ee07887
+size 1206120

data/nell.validation.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c44da469d07e6c45d8bd71cf84b8abc33b0d4ecf41d395c563f055db58ed0919
-size 222265
+oid sha256:4ec1b1d9c96142c349d28ae381357ed34a98e53b2e88bc4f77e7016e7545c27c
+size 132754

data/nell.vocab.clean.txt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:af38d1a58200cf9a3aa0c0d5e76c30202eff02c0cbd353f1c92640fd3fb0d422
-size 884124

data/nell.vocab.raw.txt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:29af9862df4777ce613d72e94c460ade35e60477704fa7a48274cc803ca7ea4f
-size 2114701

data/nell_filter.test.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7080aa24a120c206d15c89bdeb659613f98943e1a777dc0470d2ba220789f537
-size 328006
+oid sha256:2f8255ce0bb2c77c8a82fbb561ba299189acd79c99f55d3d4beb1a30da98d023
+size 197545

data/nell_filter.train.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5c399247378d26aaf06b3d6bc4770eb861805527311b188aba9a9243b45b84b
-size 1303312
+oid sha256:966b6dda11cec02a0b98762cd4eb277bf6a7f46eaf1f2bc4ff3a23b09a0c6a94
+size 780354

data/nell_filter.validation.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95da57e7d4674775756c8fdd3e262c1aebb094c65e4c0596a0ae0e859bf38c43
-size 190916
+oid sha256:bb7d3c43f4a4558f0ae8a3908b025cdcc889de28af65441f24afacde5522169c
+size 114372

data/wiki.test.jsonl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e16fa1f1e7a2e0b6987ba4cde8944771b5b9bd3465f8b872b2cdc324eacd05f0
-size 2044049

data/wiki.train.jsonl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:56b7715a33bdcaced2fbfdb22784d172840c9f2c21fd168defd2e79cff93bd15
-size 8154793

data/wiki.validation.jsonl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:200139f320505ba16456a9b2c6e0b9eefd65c111fb990c9a06b6ea66d4d138bf
-size 883589

data/wiki.vocab.raw.txt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:8491e7a8cd96a74c74667adfa3a8403d6040a1aaf265f8b8d2161fb2c684d119
-size 46168647

data/wiki.vocab.txt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:8491e7a8cd96a74c74667adfa3a8403d6040a1aaf265f8b8d2161fb2c684d119
-size 46168647

generate_filtered_data.py

@@ -1,87 +0,0 @@
-import json
-
-
-non_entity_types = [
-    'academicfield',
-    'agent',
-    'agriculturalproduct',
-    'amphibian',
-    'animal',
-    'aquarium',
-    'arachnid',
-    'architect',
-    'arthropod',
-    'bakedgood',
-    'bathroomitem',
-    'bedroomitem',
-    'beverage',
-    'bird',
-    'blog',
-    'bodypart',
-    'bone',
-    'candy',
-    'cave',
-    'chemical',
-    'clothing',
-    'coffeedrink',
-    'condiment',
-    'crimeorcharge',
-    'crustacean',
-    'date',
-    'dateliteral',
-    'economicsector',
-    'fish',
-    'food',
-    'fruit',
-    'fungus',
-    'furniture',
-    'grain',
-    'hallwayitem',
-    'hobby',
-    'insect',
-    'invertebrate',
-    'jobposition',
-    'kitchenitem',
-    'landscapefeatures',
-    'legume',
-    'location',
-    'mammal',
-    'meat',
-    'mlsoftware',
-    'mollusk',
-    'month',
-    'nut',
-    'officebuildingroom',
-    'physiologicalcondition',
-    'plant',
-    'politicsissue',
-    'profession',
-    'professionalorganization',
-    'reptile',
-    'room',
-    'sport',
-    'tableitem',
-    'tradeunion',
-    'vegetable',
-    'vehicle',
-    'vertebrate',
-    'weapon',
-    'wine'
-]
-
-full_data = {}
-for s in ["train", "validation", "test"]:
-    with open(f"data/nell.{s}.jsonl") as f:
-        data = [json.loads(i) for i in f.read().split('\n') if len(i) > 0]
-        data = [i for i in data if i['head_type'] not in non_entity_types and i['tail_type'] not in non_entity_types]
-    with open(f"data/nell_filter.{s}.jsonl", "w") as f:
-        f.write('\n'.join([json.dumps(i) for i in data]))
-
-
-with open("data/nell.vocab.txt") as f:
-    vocab = [i.split("\t") for i in f.read().split('\n')]
-
-vocab = ["\t".join([a, b]) for a, b in vocab if b not in non_entity_types]
-with open("data/nell_filter.vocab.txt", 'w') as f:
-    f.write('\n'.join(vocab))
-

nell.py

@@ -5,7 +5,7 @@ import datasets
 logger = datasets.logging.get_logger(__name__)
 _DESCRIPTION = """Few shots link prediction dataset. """
 _NAME = "nell"
-_VERSION = "0.0.6"
+_VERSION = "0.0.7"
 _CITATION = """
 @inproceedings{xiong-etal-2018-one,
     title = "One-Shot Relational Learning for Knowledge Graphs",
@@ -28,12 +28,6 @@ _CITATION = """
 
 _HOME_PAGE = "https://github.com/asahi417/relbert"
 _URL = f'https://huggingface.co/datasets/relbert/{_NAME}/resolve/main/data'
-# _TYPES = ["nell_filter", "nell", "wiki"]
-# _URLS = {i: {
-#     str(datasets.Split.TRAIN): [f'{_URL}/{i}.train.jsonl'],
-#     str(datasets.Split.VALIDATION): [f'{_URL}/{i}.validation.jsonl'],
-#     str(datasets.Split.TEST): [f'{_URL}/{i}.test.jsonl']
-# } for i in _TYPES}
 _TYPE = "nell_filter"
 _URLS = {
     str(datasets.Split.TRAIN): [f'{_URL}/{_TYPE}.train.jsonl'],

process.py

@@ -15,7 +15,6 @@ import re
 from itertools import chain
 
 data_dir_nell = "NELL"
-data_dir_wiki = "Wiki"
 os.makedirs("data", exist_ok=True)
 
 short = ['alcs', "uk", "us", "usa", "npr", "nbc", "bbc", "cnn", "abc", "cbs", "nfl", "mlb", "nba", "nhl", "pga", "ncaa",
@@ -114,19 +113,11 @@ if not os.path.exists(data_dir_nell):
                      "wget https://sites.cs.ucsb.edu/~xwhan/datasets/nell.tar.gz\n"
                      "tar -xzf nell.tar.gz")
 
-if not os.path.exists(data_dir_wiki):
-    raise ValueError("Please download the dataset first\n"
-                     "wget https://sites.cs.ucsb.edu/~xwhan/datasets/wiki.tar.gz\n"
-                     "tar -xzf wiki.tar.gz")
-
 
 def read_file(_file):
     with open(_file, 'r') as f_reader:
         tmp = json.load(f_reader)
     flatten = list(chain(*[[{"relation": r, "head": h, "tail": t} for (h, r, t) in v] for v in tmp.values()]))
-    # flatten = {}
-    # for k, v in tmp.items():
-    #     flatten[k] = [{"relation": r, "head": h, "tail": t} for (h, r, t) in v]
     return flatten
 
 
@@ -137,37 +128,47 @@ def read_vocab(_file):
 
 
 if __name__ == '__main__':
+    # Process raw data
     vocab = read_vocab(f"{data_dir_nell}/ent2ids")
-    with open("data/nell.vocab.raw.txt", 'w') as f:
-        f.write("\n".join(vocab))
     vocab = [clean(i) for i in vocab if len(i.split(":")) > 2]
     vocab = ["\t".join(i) for i in vocab if len(i[0]) > 0 and len(i[1]) > 0]
     with open("data/nell.vocab.txt", 'w') as f:
         f.write("\n".join(vocab))
-
-    vocab = read_vocab(f"{data_dir_wiki}/ent2ids")
-    with open("data/wiki.vocab.raw.txt", 'w') as f:
-        f.write("\n".join(vocab))
+    vocab_term = [i.split('\t')[0] for i in vocab]
 
     for i, s in zip(['dev_tasks.json', 'test_tasks.json', 'train_tasks.json'], ['validation', 'test', 'train']):
         d = read_file(f"{data_dir_nell}/{i}")
+
         for _d in d:
-            head_entity, head_type = clean(_d['head'])
-            _d['head_entity'] = head_entity
+            head = _d.pop("head")
+            tail = _d.pop("tail")
+
+            head_entity, head_type = clean(head)
+            _d['head'] = head_entity
             _d['head_type'] = head_type
+            assert head_entity in vocab_term, head_entity
 
-            tail_entity, tail_type = clean(_d['tail'])
-            _d['tail_entity'] = tail_entity
+            tail_entity, tail_type = clean(tail)
+            _d['tail'] = tail_entity
             _d['tail_type'] = tail_type
+            assert tail_entity in vocab_term, tail_entity
 
         with open(f"data/nell.{s}.jsonl", "w") as f:
             f.write("\n".join([json.dumps(_d) for _d in d]))
 
-        d = read_file(f"{data_dir_wiki}/{i}")
-        for _d in d:
-            _d['head_entity'] = ''
-            _d['head_type'] = ''
-            _d['tail_entity'] = ''
-            _d['tail_type'] = ''
-        with open(f"data/wiki.{s}.jsonl", "w") as f:
-            f.write("\n".join([json.dumps(_d) for _d in d]))
+    # Filter entity relation
+    full_data = {}
+    for s in ["train", "validation", "test"]:
+        with open(f"data/nell.{s}.jsonl") as f:
+            data = [json.loads(i) for i in f.read().split('\n') if len(i) > 0]
+            data = [i for i in data if i['head_type'] not in non_entity_types and i['tail_type'] not in non_entity_types]
+        with open(f"data/nell_filter.{s}.jsonl", "w") as f:
+            f.write('\n'.join([json.dumps(i) for i in data]))
+
+    with open("data/nell.vocab.txt") as f:
+        vocab = [i.split("\t") for i in f.read().split('\n')]
+
+    vocab = ["\t".join([a, b]) for a, b in vocab if b not in non_entity_types]
+    with open("data/nell_filter.vocab.txt", 'w') as f:
+        f.write('\n'.join(vocab))
+

stats/stats.test.entity.csv

@@ -1,64 +0,0 @@
-entity_type,nell (head),nell_filter (head),nell (tail),nell_filter (tail)
-politicianus,352,352,360,360
-sportsteam,295,295,0,0
-automobilemaker,274,273,54,54
-insect,230,0,270,0
-automobilemodel,100,100,0,0
-animal,97,0,30,0
-sport,93,0,74,0
-agriculturalproduct,87,0,0,0
-sportsgame,74,0,0,0
-product,62,62,0,0
-software,42,42,0,0
-city,42,42,161,161
-stateorprovince,38,38,0,0
-athlete,34,0,59,59
-arthropod,32,0,41,0
-organization,32,32,2,2
-beverage,27,0,0,0
-country,27,27,317,91
-geopoliticallocation,24,24,14,8
-mammal,23,0,0,0
-politician,23,23,58,58
-person,14,0,96,96
-invertebrate,14,0,43,0
-personmexico,13,0,20,20
-coffeedrink,11,0,0,0
-school,11,11,0,0
-crustacean,11,0,25,0
-county,10,10,4,4
-hobby,10,0,0,0
-vegetable,8,0,0,0
-personaustralia,4,0,5,5
-videogame,4,4,0,0
-reptile,4,0,0,0
-celebrity,4,4,2,2
-food,4,0,0,0
-male,3,1,5,5
-visualizablescene,3,3,3,3
-female,3,3,3,3
-vehicle,2,0,0,0
-amphibian,2,0,0,0
-grain,2,0,0,0
-arachnid,1,0,6,0
-drug,1,1,0,0
-personus,1,1,6,6
-professor,1,1,0,0
-legume,1,0,0,0
-coach,1,0,245,245
-chemical,1,0,0,0
-company,1,1,147,144
-mlsoftware,1,0,0,0
-island,1,1,0,0
-geopoliticalorganization,1,1,17,7
-personnorthamerica,1,0,3,3
-fruit,1,0,0,0
-planet,0,0,1,1
-biotechcompany,0,0,11,10
-bodypart,0,0,69,0
-location,0,0,2,0
-journalist,0,0,1,1
-astronaut,0,0,1,1
-director,0,0,1,1
-personeurope,0,0,1,1
-criminal,0,0,1,1

stats/stats.test.relation.csv

@@ -1,12 +0,0 @@
-relation_type,nell,nell_filter
-concept:animalsuchasinvertebrate,415,0
-concept:politicianusendorsespoliticianus,386,386
-concept:teamcoach,341,341
-concept:producedby,213,209
-concept:automobilemakerdealersincity,178,177
-concept:geopoliticallocationresidenceofpersion,143,143
-concept:agriculturalproductcamefromcountry,140,0
-concept:sportschoolincountry,103,0
-concept:automobilemakerdealersincountry,96,96
-concept:sportsgamesport,74,0
-concept:athleteinjuredhisbodypart,69,0

stats/stats.train.entity.csv

@@ -1,150 +0,0 @@
-entity_type,nell (head),nell_filter (head),nell (tail),nell_filter (tail)
-insect,782,0,904,0
-country,774,755,885,455
-ceo,433,423,1,0
-politicianus,416,408,25,12
-sportsteam,392,392,430,430
-academicfield,388,0,377,0
-sportsleague,356,356,12,12
-athlete,353,353,22,21
-city,352,342,864,852
-person,351,350,296,256
-weapon,339,0,0,0
-stateorprovince,263,254,736,602
-animal,245,0,23,0
-geopoliticallocation,195,184,135,112
-agriculturalproduct,170,0,56,0
-food,163,0,58,0
-airport,152,152,0,0
-sport,139,0,0,0
-male,132,132,81,78
-automobilemaker,131,131,29,29
-musicartist,121,118,5,5
-female,117,116,8,8
-bank,109,109,126,126
-musicgenre,107,107,107,107
-politician,107,107,10,5
-arthropod,103,0,86,0
-drug,92,91,0,0
-crimeorcharge,83,0,0,0
-vegetable,81,0,145,0
-company,79,76,549,549
-mammal,72,0,31,0
-personmexico,57,57,16,14
-school,54,54,1,1
-clothing,52,0,0,0
-crustacean,46,0,31,0
-personus,42,41,25,21
-transportation,38,36,2,2
-personaustralia,38,38,6,5
-county,36,36,39,39
-candy,34,0,0,0
-geopoliticalorganization,32,28,90,68
-beverage,30,0,20,0
-grain,29,0,6,0
-coach,29,29,62,61
-invertebrate,28,0,8,0
-arachnid,25,0,17,0
-governmentorganization,25,25,95,95
-organization,24,23,87,86
-fish,21,0,75,0
-visualizablescene,20,20,7,7
-personcanada,19,19,15,14
-meat,19,0,0,0
-island,15,15,4,4
-hobby,15,0,0,0
-biotechcompany,14,14,80,80
-economicsector,12,0,14,0
-bathroomitem,11,0,0,0
-website,10,7,32,31
-personnorthamerica,10,9,7,6
-personeurope,9,9,8,7
-professor,7,7,2,2
-port,7,7,0,0
-university,6,3,15,15
-celebrity,6,6,5,5
-actor,6,6,3,2
-scientist,5,5,2,2
-wine,5,0,4,0
-musician,5,5,124,124
-bakedgood,4,0,0,0
-bone,4,0,0,0
-attraction,4,4,2,1
-astronaut,4,4,0,0
-coffeedrink,4,0,0,0
-building,4,4,2,0
-monarch,4,4,3,3
-journalist,4,4,1,0
-fruit,3,0,0,0
-newspaper,3,3,2,2
-bird,3,0,22,0
-vertebrate,3,0,18,0
-criminal,3,3,1,0
-amphibian,3,0,6,0
-winery,2,0,0,0
-director,2,2,0,0
-mollusk,2,0,0,0
-bedroomitem,2,0,0,0
-politicsblog,2,2,6,3
-kitchenitem,2,0,0,0
-visualizablething,2,1,3,1
-model,2,2,0,0
-bodypart,1,0,0,0
-visualizableobject,1,0,1,0
-politicsissue,1,0,10,0
-recordlabel,1,1,13,13
-personafrica,1,1,4,3
-continent,1,0,1,1
-furniture,1,0,0,0
-reptile,1,0,12,0
-retailstore,1,1,15,15
-personasia,1,1,3,3
-location,1,0,6,0
-fungus,1,0,0,0
-writer,1,0,6,3
-creditunion,1,1,0,0
-condiment,1,0,0,0
-hallwayitem,1,0,0,0
-comedian,1,1,1,0
-tableitem,1,0,0,0
-publication,1,1,21,21
-planet,1,1,0,0
-museum,1,1,5,5
-personsouthamerica,1,1,1,1
-legume,0,0,6,0
-petroleumrefiningcompany,0,0,6,6
-tradeunion,0,0,2,0
-televisionnetwork,0,0,1,1
-nut,0,0,1,0
-blog,0,0,17,0
-disease,0,0,269,92
-architect,0,0,1,0
-profession,0,0,1,0
-televisionstation,0,0,221,221
-nongovorganization,0,0,5,4
-physiologicalcondition,0,0,2,0
-cave,0,0,1,0
-zoo,0,0,1,1
-caf_,0,0,1,1
-visualartist,0,0,1,0
-magazine,0,0,5,5
-nonprofitorganization,0,0,2,2
-hotel,0,0,3,1
-park,0,0,1,1
-professionalorganization,0,0,9,0
-plant,0,0,82,0
-landscapefeatures,0,0,8,0
-aquarium,0,0,1,0
-trainstation,0,0,2,2
-agent,0,0,1,0
-month,0,0,48,0
-date,0,0,2,0
-politicalparty,0,0,6,6
-radiostation,0,0,93,93
-officebuildingroom,0,0,6,0
-jobposition,0,0,5,0
-river,0,0,4,4
-dateliteral,0,0,1,0
-stadiumoreventvenue,0,0,417,417
-room,0,0,10,0
-politicaloffice,0,0,216,216

stats/stats.train.relation.csv

@@ -1,52 +0,0 @@
-relation_type,nell,nell_filter
-concept:athleteledsportsteam,424,424
-concept:academicfieldsuchasacademicfield,401,0
-concept:topmemberoforganization,364,354
-concept:teamplaysincity,338,338
-concept:citytelevisionstation,316,316
-concept:animalsuchasinsect,291,0
-concept:hasofficeincountry,283,283
-concept:leaguestadiums,279,279
-concept:ceoof,276,271
-concept:politicianrepresentslocation,260,258
-concept:weaponmadeincountry,257,0
-concept:animalthatfeedoninsect,238,0
-concept:personmovedtostateorprovince,225,225
-concept:arthropodandotherarthropod,222,0
-concept:politicianusholdsoffice,222,216
-concept:countrycapital,213,211
-concept:airportincity,210,210
-concept:inverseofarthropodcalledarthropod,199,0
-concept:countryhascitizen,183,182
-concept:countryoforganizationheadquarters,181,166
-concept:animaleatvegetable,178,0
-concept:countrystates,169,169
-concept:sportfansincountry,154,0
-concept:statehascapital,151,151
-concept:teamhomestadium,138,138
-concept:invertebratefeedonfood,136,0
-concept:agriculturalproductcomingfromvertebrate,125,0
-concept:fooddecreasestheriskofdisease,122,1
-concept:personleadsgeopoliticalorganization,120,120
-concept:stateorprovinceoforganizationheadquarters,120,118
-concept:musicartistmusician,118,118
-concept:itemexistsatlocation,110,0
-concept:fatherofperson,108,108
-concept:musicgenressuchasmusicgenres,107,107
-concept:countriessuchascountries,100,100
-concept:cityradiostation,99,99
-concept:wifeof,99,99
-concept:agriculturalproducttoattractinsect,97,0
-concept:drugpossiblytreatsphysiologicalcondition,92,91
-concept:crimeorchargeofperson,83,0
-concept:agriculturalproductgrowinginstateorprovince,79,0
-concept:automobilemakercardealersinstateorprovince,78,78
-concept:personalsoknownas,78,78
-concept:animalsuchasfish,74,0
-concept:leaguecoaches,71,71
-concept:organizationnamehasacronym,61,61
-concept:bankboughtbank,59,58
-concept:foodcancausedisease,57,0
-concept:vegetableproductioninstateorprovince,56,0
-concept:clothingmadefromplant,54,0
-concept:organizationdissolvedatdate,51,0

stats/stats.validation.entity.csv

@@ -1,32 +0,0 @@
-entity_type,nell (head),nell_filter (head),nell (tail),nell_filter (tail)
-city,316,316,316,316
-bank,144,144,0,0
-sport,125,0,0,0
-person,116,116,131,131
-geopoliticallocation,96,96,29,29
-governmentorganization,74,74,0,0
-female,38,38,9,9
-male,37,37,52,52
-personeurope,14,14,4,4
-county,11,11,11,11
-geopoliticalorganization,8,8,21,21
-monarch,4,4,1,1
-island,4,4,6,6
-politicianus,3,3,71,71
-visualizablescene,3,3,3,3
-personus,2,2,0,0
-politicalparty,2,2,0,0
-writer,1,1,0,0
-room,1,0,0,0
-visualizablething,1,1,1,1
-organization,1,1,1,1
-criminal,1,1,0,0
-company,1,1,0,0
-athlete,1,1,2,2
-scientist,0,0,1,1
-coach,0,0,3,3
-astronaut,0,0,1,1
-politician,0,0,1,1
-stadiumoreventvenue,0,0,125,0
-personsouthamerica,0,0,17,17
-country,0,0,198,197

stats/stats.validation.relation.csv

@@ -1,6 +0,0 @@
-relation_type,nell,nell_filter
-concept:cityalsoknownas,356,356
-concept:bankbankincountry,230,229
-concept:parentofperson,217,217
-concept:sportusesstadium,125,0
-concept:politicalgroupofpoliticianus,76,76

stats/stats.vocab.csv

@@ -1,280 +0,0 @@
-entity_type,nell,nell_filter,sample
-person,4026,4026,"George P Schultz, Arthur, Les Miserables"
-book,3835,3835,"The Heart Of The Matter, Leadership, Battlefield Earth"
-city,3020,3020,"Kyoto, Maryville, Name"
-athlete,2965,2965,"Linas Kleiza, Josh Butler, Jose Acevedo"
-company,2520,2520,"Dialogic, Kfqx, Kdse"
-sportsteam,1799,1799,"Crew, Winston Salem State Rams, Se Missouri State Indians"
-clothing,1589,0,"calico, mini skirt, bag"
-writer,1345,1345,"Max Frisch, Michael J Fox, Joyce Grenfell"
-academicfield,1154,0,"resource management, wellness, communications studies"
-drug,1030,1030,"Diltiazem, Tambocor, Trazodone"
-televisionstation,1018,1018,"Wutb, Wfft Tv, Wvny Tv"
-geopoliticallocation,1016,1016,"Baltic States, Macau, Sides"
-personus,996,996,"Gore Vidal, Khaled Hosseini, Tommy Shaw"
-weapon,932,0,"gas imports, battle ready swords, pulse rifles"
-coach,924,924,"Shanahan, Jeff Turner, Washington Redskins"
-journalist,875,875,"Mara Liasson, Patient, Scott Bordow"
-room,848,0,"study, open plan bedroom, second living area"
-musicartist,814,814,"Howard, Black Uhuru, Michael Schenker"
-politicianus,713,713,"Kevin Johnson, Dianne Feinstein, Richard Daley"
-stadiumoreventvenue,648,648,"Coliseum, Scottrade Center, Nippert Stadium"
-officebuildingroom,636,0,"one bedroom cottages, third guest bedroom, deluxe master bath"
-movie,620,620,"City Of God, Rope, Halloween"
-chemical,620,0,"atmospheric pollution, alkaline, oil"
-ceo,619,619,"Kerry Killinger, Herbert Hainer, Use Send"
-musician,619,619,"Tom Morello, Elgar, Steve Lacy"
-university,616,616,"Leiden University, Fiu, Hebrew Union College Jewish Institute Of Religion"
-geopoliticalorganization,612,612,"Pacific Grove, Waxahachie, Ibaraki Prefecture"
-stateorprovince,600,600,"Himachal Pradesh, Kashmir, Massachussetts"
-profession,592,0,"neurosurgeons, professional staff, sales"
-plant,568,0,"ancient oak, water oaks, skunk cabbage"
-aquarium,535,0,"saarbrucken zoo, turtle aquarium, 500 gallon aquarium"
-website,533,533,"Www Philly Com Inquirer, Memory, News Search"
-county,522,522,"Korea University, Boise City, Hopkinton"
-disease,510,510,"Headaches, Ain, Illness"
-male,484,484,"Library, Rick Davis, Gym Class Heroes"
-radiostation,476,476,"Wclj, Kvct, Knbr"
-country,472,472,"Netherland, Bfpo Addresses, Top States"
-jobposition,458,0,"ballerina, senior associate, economist"
-agriculturalproduct,449,0,"oilseeds, butter, mint leaves"
-biotechcompany,435,435,"Elan Pharmaceuticals, Bache Halsey Stuart Shields, Agouron Pharmaceutical"
-bank,430,430,"Bayern, Wachovia Securities, Cs First Boston"
-bakedgood,430,0,"sponge cake, puddings, hay"
-hotel,428,428,"Sands, Holiday Inn Express, London House"
-organization,424,424,"Orbimed Advisors, Protection The Department, General Electric Co"
-actor,418,418,"John Travolta, Sally Field, Uma Thurman"
-governmentorganization,409,409,"States Supreme Court, California Department, Democratic National Committee"
-personeurope,402,402,"Noel Streatfield, Pat Robertson, Jeffrey Archer"
-food,393,0,"mix, sausages, sucker"
-female,390,390,"Hera, Mary, Myla Goldberg"
-visualizablething,386,386,"Pesto, Bentley Gt, Resource Revenues"
-personcanada,365,365,"Michael Hyatt, Dan Harris, Self"
-visualizablescene,357,357,"Birthplace, Hyannis, Contest"
-newspaper,357,357,"Business News, Krqe Tv, New York Times"
-river,344,344,"Rogue River, Coco River, Mandovi"
-politician,342,342,"Alan Keyes, Aid, Representative George Miller"
-attraction,339,339,"Secred Garden, Kimmel Center, La Aurora International Airport"
-politicsblog,335,335,"Daily Record, New Statesman, Service Corps Of Retired Executives"
-animal,312,0,"pet ferrets, goat, racoons"
-celebrity,307,307,"Seton Hall Pirates, Wilson Pickett, Harvard Crimson"
-blog,306,0,"christian broadcasting network, cnet, sunday express"
-building,304,304,"Las Vegas Hilton, White House, Baker Berry Library"
-mammal,304,0,"1 million children, boomtown rats, buffalo"
-director,299,299,"Quentin Tarantino, John Schlesinger, Vincenzo Natali"
-bodypart,299,0,"organ systems, system, genital area"
-scientificterm,294,294,"Conditional Expectation, Probabilistic Pca, Support Vector Regression"
-personmexico,274,274,"Ollie Linton, Justin Bass, Fernando Cabrera"
-island,270,270,"Santa Cruz Del Islote, Obwalden, Joint Chiefs"
-personaustralia,263,263,"Brent Johnson, Malcolm Lowry, David Muir"
-hobby,258,0,"winter sports, writers, skydiving"
-transportation,258,258,"Fare, Seatac Airport, Newark International Airport"
-musicalbum,257,257,"Martin, Years, Compilation"
-park,247,247,"Rides, Dulwich, Salford Quays"
-criminal,243,243,"Mark Bryan, John Brown, Carlinhos Beira Mar"
-visualartist,239,239,"Diego, Georges Seurat, Francis Ford Coppola"
-airport,236,236,"Air Canada, Eleftherios Venizelos, Portland International"
-professor,235,235,"Nancy Thomas, David Gest, James M Cain"
-personnorthamerica,235,235,"David M Herszenhorn, Rediff, Robert F Wagner"
-bathroomitem,235,0,"soaking tub, tub upstairs, lock"
-emotion,230,230,"Love, Charity, Defiance"
-school,229,229,"Southern Utah University, Rensselaer Polytechnic Institute, Berea College"
-sportsgame,226,226,"Alds, 1999 World Series, 1982 World Series"
-buildingfeature,216,216,"Change, Large Windows, Improvement"
-insect,214,0,"leafminers, mealworms, chopin"
-furniture,210,0,"large double bed, settee, sets glass top dining sets"
-automobilemodel,210,210,"Sunliner, Edge, Taurus X"
-publication,205,205,"Financial Mail, Data Domain, Industry Standard"
-scientist,204,204,"Calvin, Klaus, Thomas Kuhn"
-magazine,203,203,"Fitness, Wwd, Southern Living"
-bird,199,0,"old world flycatchers, migratory waterfowl, black widow"
-product,194,194,"Samsung Electronics Co Ltd, Microsoft Windows Ce, Illustrator"
-professionalorganization,192,0,"maryland division, restaurant and catering association, science foundation"
-musicsong,186,186,"Constellation, Download, Any Way You Want Me"
-personafrica,185,185,"Caterpillar, Margaret Wise Brown, Anna Politkovskaya"
-automobilemaker,177,177,"Honda, Porsche, Used Car Search"
-televisionshow,176,176,"Mission, Jail, Prime Time"
-lake,172,172,"Lake Granby, Smith Mountain Lake, Lake Pend Oreille"
-museum,168,168,"New York Stock Exchange, Art, California Science Center"
-artery,167,167,"Vascular System, Pulmonary Valve, Major Artery"
-bacteria,165,165,"Echinococcus Multilocularis, Borrelia Recurrentis, Clostridium Botulinum"
-beverage,164,0,"sophisticated water, soymilk, mixer"
-language,162,162,"Gujarati, Creole, Kinyarwanda"
-comedian,159,159,"Warren Zevon, Clive Barker, Bob Young"
-politicsissue,152,0,"fiscal policy, resource services, health services"
-mountain,150,150,"Diamond Head, Pleasant Ridge, Great Smoky Mountain"
-landscapefeatures,148,0,"green pastures, hot water, renaissance"
-fish,146,0,"starfish, tilapia, pelagic fish"
-retailstore,138,138,"U S Post Office, Mcgraw Hill, Sam Walton"
-sportsequipment,138,138,"West, Instructions, Ball"
-musicinstrument,135,135,"Rock Band, Fortepiano, Menu"
-skiarea,132,132,"Forked River, Minnesota River, Alta"
-programminglanguage,131,131,"Enter, Montpellier, Site"
-software,130,130,"Adobe Golive, Nemo, Excel 2"
-winery,126,126,"Legacy, Rockland Farms Winery, Harvest"
-geometricshape,125,125,"Units, Rates, Cone"
-sport,123,0,"swimming, fantasy football, snorkelling"
-location,122,0,"home water, representative, ideal location"
-physiologicalcondition,121,0,"medical illness, male impotence, chronic arthritis"
-recordlabel,121,121,"Blue Note, Top Rank, Emi Group"
-vertebrate,120,0,"sugar gliders, animals people, tyrant flycatchers"
-physicalaction,115,115,"Purpose, Share, Fusion"
-personasia,115,115,"Chris Volstad, Gauri Khan, Ian Kinsler"
-musicgenre,107,107,"Glam Metal, British Folk, Power Pop"
-vehicle,103,0,"pumps, forklifts, animals"
-sportsteamposition,102,102,"Defensive Tackle, Hitters, League"
-date,101,0,"republic last year, the early 1990s, leonardo"
-eventoutcome,100,100,"Outcome, Triumph, Sales"
-vegetable,99,0,"peas, sweet potato, carrot"
-sportsleague,98,98,"Partners, Sports, Uefa"
-arthropod,96,0,"buckeye, verruca stroemia, tiny parasites"
-awardtrophytournament,95,95,"Grand Slam, U S A, Awards"
-politicalparty,95,95,"Page Content, Conservative Republicans, Rival Fatah Movement"
-politicaloffice,95,95,"Co Chair, Bill, United States Presidency"
-hallwayitem,93,0,"brand, foundation, fall"
-model,92,92,"Dan Rather, Gabriel Garc A M Rquez, Dior"
-shoppingmall,91,91,"Circus Circus, Real Place, Shares"
-terroristorganization,87,87,"Chins, Palestinian Leadership, People S Liberation Army"
-hospital,86,86,"Memphis, Toronto General, Memorial Sloan Kettering"
-economicsector,86,0,"motorcycle, computer, home"
-braintissue,86,86,"Genitourinary System, Experiment, Assessment"
-color,84,84,"Pink, Shade, Green"
-weatherphenomenon,83,83,"Earthquake, Clouds, Threat"
-bedroomitem,82,0,"sale, air conditioning, action"
-reptile,81,0,"arrow, short eared owl, wading birds"
-chef,80,80,"John Gatto, Len Deighton, Jeffrey Gettleman"
-currency,79,79,"American, Sierra Leone, Lloyds"
-invertebrate,79,0,"infusoria, sea life, egrets"
-amphibian,78,0,"albino fire salamander, salamanders of alabama, larval amphibians"
-monument,77,77,"Close, Various Times, Canada Aviation Museum"
-charactertrait,77,77,"Multitude, Office 2, Hoyas"
-street,76,76,"Middlesex Turnpike, Capitol Street, Important Step"
-visualizableobject,72,72,"Desktop Computer, Member, Telescope"
-restaurant,69,69,"Duck And Waffle, Oblix, The Range Steakhouse"
-wine,69,0,"burkina faso, oakville, perugia"
-nongovorganization,69,69,"Fee, Communications Committee, Moderate Fatah Party"
-buildingmaterial,68,68,"Iron, Drainage, Fibreglass"
-videogame,66,66,"Breath Of Fire 4, Charlie And The Chocolate Factory, Vessel"
-dateliteral,65,0,"1911, 1938, 2010"
-nerve,65,65,"Nerve Roots, Brain Function, Damage"
-architect,65,0,"robert t coles, hubert robert, max bond"
-monarch,59,59,"Peter Kropotkin, All Star, Pertinax"
-religion,57,57,"Firestone Park, Developments, Roman Catholic"
-videogamesystem,56,56,"Xboxone, Blackberry, System Maintenance"
-condiment,56,0,"mushroom sauce, lemon sauce, wasabi"
-trainstation,54,54,"Ideal Base, Needs, Sacramento Rt"
-mldataset,54,54,"William, Hdv, Hq"
-year,53,53,"1874, 90, Start"
-personsouthamerica,52,52,"Diesel, Gas, Iesus"
-beach,52,52,"Pine Valley, Daufuskie Island, Pensacola Beach"
-kitchenitem,52,0,"public water, copy, fridge freezer"
-astronaut,51,51,"Helen Sharman, Exploration, Guion Blufordand"
-muscle,50,50,"Dorsal Root Ganglion, Fatigue, Pectoralis Minor"
-crimeorcharge,50,0,"gun crimes, policy, abuse"
-parlourgame,50,50,"1 2 Weeks, Water, 12 Hours"
-ethnicgroup,50,50,"Burundi, Turkish, Realtime"
-zoo,49,49,"Birmingham Zoo, Ellen Trout Zoo, Aquarium Of The Americas"
-nonprofitorganization,49,49,"Data, Charity Navigator, South Central Los Angeles"
-farm,49,49,"Woodbury Farm, Farms, Cross Roads"
-consumerelectronicitem,47,47,"Gba Sp, Muse, Wii Games"
-tool,47,47,"Studs, Companies, Schedule Software"
-agent,47,0,"metaweb, french, dreyfus funds"
-mollusk,46,0,"domesticated animals, survey, sea anemones"
-port,46,46,"Southampton, Port Clinton, Andalucia"
-meat,46,0,"ham, whole chicken, ground chicken"
-mlauthor,46,46,"Carlos Brito, Web Search, Roy Williams"
-crustacean,45,0,"gulf, function, white"
-musicfestival,45,45,"Annual Event, Wolf Creek Inn State Heritage Site, Newport Music Hall"
-tableitem,43,0,"skills, dinner, laptop"
-planet,42,42,"Reasons, 10 0, Heaven"
-conference,41,41,"Filesystem, Tocs, Trsm"
-arachnid,40,0,"oklahoma, middle, distance charges"
-protein,40,40,"Proteases, Vitamin D, Effect"
-tradeunion,40,0,"streets, program, legend"
-skyscraper,40,40,"Property Types, Landmark Skyscraper, Citigroup Center"
-nondiseasecondition,39,39,"Urinary Tract Infections, Symptom, Development Corporation"
-visualartform,36,36,"Materials, Contemporary Art, Friends"
-grain,35,0,"hrs, oil, pole beans"
-visualartmovement,34,34,"Cubism, Palm Springs, United States"
-fruit,34,0,"flax seed, black cherries, pineapple"
-wallitem,33,33,"Description, East, Output"
-candy,33,0,"gathering, peanut butter and jelly fudge, starbucks coffee"
-bone,32,0,"two, ligament, chemotherapy"
-highway,30,30,"Gas, Crestline, Disability"
-election,30,30,"Republican, The National, Users"
-mlsoftware,30,0,"mallet, p g, favorite social bookmarking websites comments"
-event,29,29,"Pearl Harbour, Events Click, System Improvements"
-nonneginteger,28,28,"90, Person Time, Great Time"
-placeofworship,28,28,"Sleepy Hollow Presbyterian Church, Banteay Srei, Law Society"
-boardgame,27,27,"Life, Computer, Anagrams"
-visualizableattribute,27,27,"Dark Green Foliage, Northern States, Colour"
-politicsbill,26,26,"Pennsylvania Law, States Constitution, Activesync"
-householditem,26,26,"Brockton House Inn Bed, Home Office, Size Beds"
-creditunion,25,25,"Good Deal, Arizona, Missouri"
-perceptionaction,25,25,"Prior Approval, Mom, Waterloo"
-creativework,24,24,"Uncle Silas, Print The Application, Future U S"
-automobileengine,24,24,"Ford Ranger, Ford Five Hundred, Publishing"
-mlalgorithm,22,22,"2 0, Murcia, 8 25"
-race,22,22,"Toulouse, Dates, Morbihan"
-mlarea,22,22,"Analyses, Roles, Transactions"
-personalcareitem,22,22,"Lancaster, Body Parts, Hairdryer"
-nut,22,0,"horses, hazelnuts, heath"
-researchproject,21,21,"Wonders, Jersey Research, Consultants"
-sociopolitical,21,21,"Welcome Thing, Memoirs Of A Revolutionist, Good Idea"
-vein,21,21,"Section, Pituitary Gland, Macula"
-archaea,20,20,"Haloarcula Strains, Mesophile Ferroplasma Acidarmanus, Acidophilic Archaea"
-bridge,20,20,"Tower Bridge, Lincoln Tunnel, Life Sciences Institute"
-cheese,19,19,"Close Up, Boursin Cheese, Ricotta Cheese"
-legume,19,0,"close, wide screen, color television"
-mountainrange,17,17,"England Mountains, Smith River, Point Hope"
-physicalcharacteristic,17,17,"Possibility, Scientific Advisory, Hurst"
-sportsevent,16,16,"Sarthe, Series Championship, Premier"
-medicalprocedure,16,16,"Cpr, The Breast, Follow Up Study"
-cardgame,16,16,"Hearts, Vegas Casino, Numbers"
-militaryconflict,15,15,"Eight Years, United States Territory, War"
-mlmetric,15,15,"7, 13 5, 9 11"
-celltype,14,14,"Liver, Capacity, State"
-petroleumrefiningcompany,14,14,"Arcelormittal, Valero, Exxon Mobil Corp"
-mlconference,13,13,"Kingman, Change, Coordination"
-traditionalgame,13,13,"Betfair, 3 4, 2 1 2"
-meetingeventtitle,12,12,"Holey Grail The Mechanism Of Transport Through The Nuclear Pore Of Cells, Brain Maps To Mechanisms Neural Circuit Molecular Architecture, Rsa In The Real World"
-convention,12,12,"Space, Excaliber, Technorati"
-meetingeventtype,11,11,"Vasc Sminar, Special Sdi Seminar, Scs Author Presentation Booksigning"
-judge,11,11,"The California, Great Judge, Personnel"
-trail,11,11,"Carroll, Campground, Burlington"
-cognitiveactions,10,10,"Learning, Crisis, Three Hours"
-mediatype,10,10,"Singles, Clicks, Irrigation"
-month,10,0,"november, january, december"
-coffeedrink,10,0,"espresso, mocha, chocolate raspberry"
-zipcode,10,10,"33701, 80202, 20505"
-physicsterm,10,10,"15 Miles, 19 9 Percent, Gravity"
-perceptionevent,9,9,"Improvements, Macromedia Flash, Beep"
-cave,8,0,"will, w d, anyone"
-lymphnode,8,8,"Lymph Nodes, Enzymes, Nodes"
-filmfestival,7,7,"Igoogle, Beginning, Venice"
-fungus,7,0,"compound, enrichment, mushrooms"
-officeitem,7,7,"Oracle Corporation White Paper, Capacities, York Page"
-continent,7,7,"South America, Africa, Antarctica"
-olympics,6,6,"CBS Television Network, Palestinian National Authority, Mayor"
-url,6,6,"Links Links, Numerous Links, Google Link"
-game,6,6,"Designee The Vice, Team Plays, Aspects"
-flooritem,5,5,"Size Sleeper Sofa, Size Sofa Bed, Water Quality"
-televisionnetwork,5,5,"Cw, Upn, CNN PBS"
-mediacompany,4,4,"Vibrant Media, Demand Media, Media Public"
-dayofweek,4,4,"Monday, Tuesday, Wednesday"
-time,4,4,"4 00 P M Contact, 9 00 Pm, 7 Click"
-caf_,3,3,"Starbucks Coffee Company, Sbux, Tim Hortons"
-item,3,3,"Dishes, Laptop, Laundry Detergent"
-militaryeventtype,3,3,"Clashes, Attacks, Regimes"
-victim,3,3,"Resources Students, Technology Students, Health Sciences Students"
-geolocatablething,3,3,"Buildings, Iquique, Cars"
-grandprix,2,2,"Bernie Ecclestone, Renault"
-politicsgroup,2,2,"Description, Moveon Org"
-gamescore,2,2,"4 5, 2 3"
-refineryproduct,2,2,"East Coast Ports, Coolant"
-recipe,1,1,Rice
-virus,1,1,Safe Drinking Water
-humanagent,1,1,Monsanto Co S G D Searle Division
-species,1,1,Marine Flora
-personbylocation,1,1,Dan Garton
-SUM,68518,53887,