Datasets labelled Demographics

16 extractive question-and-answer pairs built from Report on the Census of the Madras Presidency, Vol-I, Tamilnadu, published by censusindia.gov.in. The source is a 92-page publication. Source-evidenced topics include 1871 census, madras presidency, population, enumeration, tabulation, districts, religion, castes. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 14 factual, 1 list, 1 summary. 2 of the 16 pairs (12.5%) ask explanatory questions; the remaining 14 are factual or table-based questions. 93.75% of rows pass the corpus quality gate.

44 extractive question-and-answer pairs built from Census Report, West Bengal, published by censusindia.gov.in. The source is a 360-page publication. Source-evidenced topics include census, accuracy, hurdwan division, presidency division, deces division, bhangulpore division, sonthal mexgununks, oriase division. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 36 factual, 4 summary, 2 list, 2 relationship. 8 of the 44 pairs (18.2%) ask explanatory questions; the remaining 36 are factual or table-based questions. 97.73% of rows pass the corpus quality gate.

726 extractive question-and-answer pairs built from Dh 2011 2822 Part a Dchb Anantapur, published by censusindia.gov.in. The source is a 1084-page publication. Source-evidenced topics include andhra pradesh, district census handbook, anantapur, anantasagaram river, musalammakatta, primary census abstract, village directory, town directory. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 408 factual, 133 list, 79 definition, 71 summary, 28 relationship, 7 comparison. 318 of the 726 pairs (43.8%) ask explanatory questions; the remaining 408 are factual or table-based questions. 98.76% of rows pass the corpus quality gate.

132 extractive question-and-answer pairs built from Dh 2011 1817 Part B Dchb Cachar, published by censusindia.gov.in. The source is a 292-page publication. Source-evidenced topics include assam, cachar, district census handbook, primary census abstract, kanchakanti kali mandir, silchar, kumbhirgram airport, 2011 census. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 77 factual, 25 list, 11 summary, 10 relationship, 8 definition, 1 comparison. 55 of the 132 pairs (41.7%) ask explanatory questions; the remaining 77 are factual or table-based questions. 94.70% of rows pass the corpus quality gate.

148 extractive question-and-answer pairs built from Dh 2011 1822 Part B Dchb Kamrup, published by censusindia.gov.in. The source is a 278-page publication. Source-evidenced topics include assam, kamrup, district census handbook, primary census abstract, sualkuchi, brahmaputra, muga silk, pat silk. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 90 factual, 22 list, 17 summary, 9 definition, 7 relationship, 3 comparison. 58 of the 148 pairs (39.2%) ask explanatory questions; the remaining 90 are factual or table-based questions. 93.92% of rows pass the corpus quality gate.

148 extractive question-and-answer pairs built from Dh 2011 1823 Part B Dchb Kamrup Metropolitan, published by censusindia.gov.in. The source is a 170-page publication. Source-evidenced topics include assam, kamrup metropolitan, district census handbook, primary census abstract, umananda island, guwahati, ahom king gadadhar singha, shiva temple. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 86 factual, 27 list, 17 summary, 9 relationship, 6 definition, 3 comparison. 62 of the 148 pairs (41.9%) ask explanatory questions; the remaining 86 are factual or table-based questions. 97.97% of rows pass the corpus quality gate.

207 extractive question-and-answer pairs built from Dh 2011 2820 Part B Dchb Ysr, published by censusindia.gov.in. The source is a 548-page publication. Source-evidenced topics include andhra pradesh, ysr district, district census handbook, primary census abstract, tallapaka, annamacharya, rayalaseema region, kadapa. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 128 factual, 29 list, 22 summary, 16 relationship, 9 definition, 3 comparison. 79 of the 207 pairs (38.2%) ask explanatory questions; the remaining 128 are factual or table-based questions. 96.14% of rows pass the corpus quality gate.

836 extractive question-and-answer pairs built from Dh 2011 1919 Part a Dchb Purba Medinipur, published by censusindia.gov.in. The source is a 1236-page publication. Source-evidenced topics include west bengal, purba medinipur, district census handbook, mahishadal rajbari, tamluk, 16th century, 18th century, rani janaki devi. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 444 factual, 127 list, 114 definition, 91 summary, 51 relationship, 6 comparison, 3 table. 389 of the 836 pairs (46.5%) ask explanatory questions; the remaining 447 are factual or table-based questions. 96.29% of rows pass the corpus quality gate.

282 extractive question-and-answer pairs built from Dh 2011 1918 Part B Dchb Paschim Medinipur, published by censusindia.gov.in. The source is a 1236-page publication. Source-evidenced topics include males females, persons males females, persons males, males females persons, females persons, females persons males, census abstract, primary census abstract. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 143 factual, 50 list, 37 summary, 25 relationship, 21 definition, 6 comparison. 139 of the 282 pairs (49.3%) ask explanatory questions; the remaining 143 are factual or table-based questions. 90.43% of rows pass the corpus quality gate.

837 extractive question-and-answer pairs built from Dh 2011 1917 Part a Dchb South Twenty Four Parganas, published by censusindia.gov.in. The source is a 1048-page publication. Source-evidenced topics include west bengal, south twenty four parganas, district census handbook, c.d. block, ganga sagar, royal bengal tiger, sundarban, makar sankranti. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 477 factual, 107 list, 105 summary, 96 definition, 40 relationship, 9 comparison, 3 table. 357 of the 837 pairs (42.7%) ask explanatory questions; the remaining 480 are factual or table-based questions. 97.01% of rows pass the corpus quality gate.

265 extractive question-and-answer pairs built from Dh 2011 1913 Part a Dchb Bankura, published by censusindia.gov.in. The source is a 1448-page publication. Source-evidenced topics include west bengal, district census handbook, bankura, terracotta, bankura horse, panchmura, primary census abstract, scheduled castes. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 156 factual, 49 list, 27 definition, 16 summary, 15 relationship, 2 comparison. 109 of the 265 pairs (41.1%) ask explanatory questions; the remaining 156 are factual or table-based questions. 96.98% of rows pass the corpus quality gate.

280 extractive question-and-answer pairs built from Dh 2011 1100 Part B Dchb Sikkim, published by censusindia.gov.in. The source is a 308-page publication. Source-evidenced topics include sikkim, district census handbook, north district, west district, south district, east district, teesta river, rangeet river. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 157 factual, 52 list, 29 summary, 21 definition, 20 relationship, 1 comparison. 123 of the 280 pairs (43.9%) ask explanatory questions; the remaining 157 are factual or table-based questions. 98.21% of rows pass the corpus quality gate.

812 extractive question-and-answer pairs built from Microsoft Word - DCHB Andaman - Part A workbook Version II.docx, published by censusindia.gov.in. The source is a 450-page publication. Source-evidenced topics include andaman & nicobar islands, district census handbook, cellular jail, port blair, national memorial, 1857 freedom movement, marina park, primary census abstract (pca). Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 455 factual, 119 list, 94 summary, 89 definition, 32 relationship, 18 table, 5 comparison. 339 of the 812 pairs (41.7%) ask explanatory questions; the remaining 473 are factual or table-based questions. 97.41% of rows pass the corpus quality gate.

838 extractive question-and-answer pairs built from Dh 2011 2820 Part a Dchb Ysr, published by censusindia.gov.in. The source is a 954-page publication. Source-evidenced topics include andhra pradesh, district census handbook, tallapaka, annamacharya, telugu poet, lord venkateswara, rayalaseema region, rajampet mandal. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 465 factual, 170 list, 107 definition, 59 summary, 28 relationship, 6 comparison, 3 table. 370 of the 838 pairs (44.2%) ask explanatory questions; the remaining 468 are factual or table-based questions. 97.37% of rows pass the corpus quality gate.

289 extractive question-and-answer pairs built from Dh 2011 1912 Part B Dchb Hugli, published by censusindia.gov.in. The source is a 516-page publication. Source-evidenced topics include west bengal, hugli, district census handbook, primary census abstract, bandel church, hugli river, william carey, samachar darpan. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 190 factual, 33 list, 29 summary, 18 relationship, 17 definition, 2 comparison. 99 of the 289 pairs (34.3%) ask explanatory questions; the remaining 190 are factual or table-based questions. 93.77% of rows pass the corpus quality gate.

800 extractive question-and-answer pairs built from Dh 2011 1914 Part a Dchb Puruliya, published by censusindia.gov.in. The source is a 1098-page publication. Source-evidenced topics include west bengal, puruliya, district census handbook, chhau dance, chorida, jhumur songs, national highway 32, subarnarekha river. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 411 factual, 124 list, 117 definition, 103 summary, 29 relationship, 10 comparison, 6 table. 383 of the 800 pairs (47.9%) ask explanatory questions; the remaining 417 are factual or table-based questions. 97.75% of rows pass the corpus quality gate.

892 extractive question-and-answer pairs built from Dh 2011 1915 Part a Dchb Haora, published by censusindia.gov.in. The source is a 514-page publication. Source-evidenced topics include west bengal, district census handbook, haora, total area, total population, vidyasagar setu, river ganga, hooghly. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 521 factual, 146 list, 94 definition, 82 summary, 40 relationship, 9 comparison. 371 of the 892 pairs (41.6%) ask explanatory questions; the remaining 521 are factual or table-based questions. 98.21% of rows pass the corpus quality gate.

1085 extractive question-and-answer pairs built from Dh 2011 0307 Part a Dchb Ludhiana, published by censusindia.gov.in. The source is a 430-page publication. Source-evidenced topics include punjab, district census handbook, ludhiana, district shahid bhagat singh nagar, district rupnagar, district sangrur, chandigarh, ambala. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 680 factual, 142 list, 111 definition, 86 summary, 51 relationship, 15 comparison. 405 of the 1085 pairs (37.3%) ask explanatory questions; the remaining 680 are factual or table-based questions. 95.02% of rows pass the corpus quality gate.

79 extractive question-and-answer pairs built from AR Census Act 1948, published by censusindia.gov.in. The source is a 23-page publication. Source-evidenced topics include census act, 1948, india, central government, state government, census commissioner, indian penal code, district magistrate, local authority. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 58 factual, 11 list, 7 summary, 2 definition, 1 comparison. 21 of the 79 pairs (26.6%) ask explanatory questions; the remaining 58 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

35 extractive question-and-answer pairs built from Microsoft Word - 457D5983-6A49-08A128.doc, published by censusindia.gov.in. The source is a 9-page publication. Source-evidenced topics include notification, census act, 1948, census rules, 1990, census commissioner, charge officer, enumerator, state governments, union territory administrations. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 27 factual, 4 definition, 2 list, 1 relationship, 1 summary. 8 of the 35 pairs (22.9%) ask explanatory questions; the remaining 27 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

143 extractive question-and-answer pairs built from Dh 2011 0803 Part B Dchb Bikaner, published by censusindia.gov.in. The source is a 254-page publication. Source-evidenced topics include rajasthan, bikaner, district census handbook, junagarh fort, raja rai singh, emperor akbar, chandra mahal, phool mahal. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 86 factual, 26 list, 12 definition, 11 summary, 5 comparison, 3 relationship. 57 of the 143 pairs (39.9%) ask explanatory questions; the remaining 86 are factual or table-based questions. 97.90% of rows pass the corpus quality gate.

130 extractive question-and-answer pairs built from Dh 2011 0832 Part B Dchb Udaipur, published by censusindia.gov.in. The source is a 410-page publication. Source-evidenced topics include rajasthan, udaipur, district census handbook, lake palace, jag niwas, lake pichola, rana jagat singh-ii, sisodia family. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 73 factual, 22 list, 17 summary, 8 relationship, 7 definition, 3 comparison. 57 of the 130 pairs (43.8%) ask explanatory questions; the remaining 73 are factual or table-based questions. 96.15% of rows pass the corpus quality gate.

540 extractive question-and-answer pairs built from Dh 2011 0812 Part a Dchb Jaipur, published by censusindia.gov.in. The source is a 1150-page publication. Source-evidenced topics include rajasthan, jaipur, district census handbook, amber fort, man singh, jaisingh-i, jaisingh-ii, kachahawahas. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 291 factual, 112 list, 72 definition, 33 summary, 21 relationship, 11 comparison. 249 of the 540 pairs (46.1%) ask explanatory questions; the remaining 291 are factual or table-based questions. 98.52% of rows pass the corpus quality gate.

145 extractive question-and-answer pairs built from Dh 2011 0815 Part B Dchb Jodhpur, published by censusindia.gov.in. The source is a 408-page publication. Source-evidenced topics include rajasthan, jodhpur, district census handbook, mehrangarh fort, primary census abstract, scheduled castes, scheduled tribes, village directory. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 79 factual, 30 list, 20 summary, 7 relationship, 5 definition, 4 comparison. 66 of the 145 pairs (45.5%) ask explanatory questions; the remaining 79 are factual or table-based questions. 96.55% of rows pass the corpus quality gate.