textmachine/eval/pilot/kana_traps.json

76 lines
11 KiB
JSON
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

{
"_provenance": "ja-kana precision trap corpus for kana_precision.py (D16-tail / E1-protocol extension). Sourced 2026-07-09 by a research agent: names extracted by reading the four ja PD samples (soseki-neko, akutagawa-rashomon, akutagawa-hana, dazai-merosu); collision traps from Japanese lexical knowledge, each ADJACENCY- and VOICING-checked (3 faulty candidates おなか/みがく/みんな rejected). Every trap tagged type: substring (name hides inside a longer unrelated word) vs homograph (name == a whole common noun). KEY FINDING: the substring class vanishes by MIN=3-4; the homograph class is length-INVARIANT (a MIN threshold cannot filter it — needs POS/known-word gating). kana length = CHARACTER count a substring matcher sees (ディオニス = 5 chars). Matcher kana-folds katakana->hiragana, so same-fold substrings fire.",
"names_from_corpus": [
{"name_kana":"メロス","len":3,"work":"merosu","romaji":"Merosu","kind":"person(protagonist)","written_kana_in_source":true},
{"name_kana":"ゼウス","len":3,"work":"merosu","romaji":"Zeusu","kind":"deity","written_kana_in_source":true},
{"name_kana":"シラクス","len":4,"work":"merosu","romaji":"Shirakusu","kind":"place","written_kana_in_source":true},
{"name_kana":"アレキス","len":4,"work":"merosu","romaji":"Arekisu","kind":"person(minor)","written_kana_in_source":true},
{"name_kana":"ディオニス","len":5,"work":"merosu","romaji":"Dionis","kind":"person(antagonist)","written_kana_in_source":true},
{"name_kana":"セリヌンティウス","len":8,"work":"merosu","romaji":"Serinuntiusu","kind":"person","written_kana_in_source":true},
{"name_kana":"フィロストラトス","len":8,"work":"merosu","romaji":"Firosutoratosu","kind":"person(minor)","written_kana_in_source":true},
{"name_kana":"おさん","len":3,"work":"neko","romaji":"Osan","kind":"person(maid)","written_kana_in_source":true},
{"name_kana":"ギボン","len":3,"work":"neko","romaji":"Gibon","kind":"reference(person)","written_kana_in_source":true},
{"name_kana":"ニコラス","len":4,"work":"neko","romaji":"Nikorasu","kind":"reference(person)","written_kana_in_source":true},
{"name_kana":"ハリソン","len":4,"work":"neko","romaji":"Harison","kind":"reference(person)","written_kana_in_source":true},
{"name_kana":"アンドレア","len":5,"work":"neko","romaji":"Andorea","kind":"reference(person)","written_kana_in_source":true}
],
"collision_traps": [
{"name_kana":"リン","len":2,"romaji":"Rin","containing_word_kana":"リンゴ","containing_word_meaning":"apple","type":"substring","confidence":"high"},
{"name_kana":"アイ","len":2,"romaji":"Ai","containing_word_kana":"アイス","containing_word_meaning":"ice/ice cream","type":"substring","confidence":"high"},
{"name_kana":"ラン","len":2,"romaji":"Ran","containing_word_kana":"ランチ","containing_word_meaning":"lunch","type":"substring","confidence":"high"},
{"name_kana":"はる","len":2,"romaji":"Haru","containing_word_kana":"はるばる","containing_word_meaning":"all the way/from afar","type":"substring","confidence":"high","grounded":"merosu"},
{"name_kana":"かな","len":2,"romaji":"Kana","containing_word_kana":"かなしい","containing_word_meaning":"sad","type":"substring","confidence":"high"},
{"name_kana":"なお","len":2,"romaji":"Nao","containing_word_kana":"すなお","containing_word_meaning":"obedient/honest","type":"substring","confidence":"high"},
{"name_kana":"みか","len":2,"romaji":"Mika","containing_word_kana":"みかん","containing_word_meaning":"mandarin orange","type":"substring","confidence":"high"},
{"name_kana":"ゆき","len":2,"romaji":"Yuki","containing_word_kana":"ゆきだるま","containing_word_meaning":"snowman","type":"substring","confidence":"high"},
{"name_kana":"まり","len":2,"romaji":"Mari","containing_word_kana":"あまり","containing_word_meaning":"remainder/too much","type":"substring","confidence":"high"},
{"name_kana":"かい","len":2,"romaji":"Kai","containing_word_kana":"せかい","containing_word_meaning":"world","type":"substring","confidence":"high"},
{"name_kana":"れい","len":2,"romaji":"Rei","containing_word_kana":"きれい","containing_word_meaning":"pretty/clean","type":"substring","confidence":"high"},
{"name_kana":"ゆう","len":2,"romaji":"Yuu","containing_word_kana":"ゆうがた","containing_word_meaning":"evening","type":"substring","confidence":"high","grounded":"rashomon,neko"},
{"name_kana":"なな","len":2,"romaji":"Nana","containing_word_kana":"バナナ","containing_word_meaning":"banana","type":"substring","confidence":"high"},
{"name_kana":"さや","len":2,"romaji":"Saya","containing_word_kana":"ささやく","containing_word_meaning":"to whisper","type":"substring","confidence":"high","grounded":"hana,merosu"},
{"name_kana":"まい","len":2,"romaji":"Mai","containing_word_kana":"まいにち","containing_word_meaning":"every day","type":"substring","confidence":"high"},
{"name_kana":"ちか","len":2,"romaji":"Chika","containing_word_kana":"ちかい","containing_word_meaning":"near","type":"substring","confidence":"high"},
{"name_kana":"みな","len":2,"romaji":"Mina","containing_word_kana":"みなと","containing_word_meaning":"harbor","type":"substring","confidence":"high"},
{"name_kana":"ゆい","len":2,"romaji":"Yui","containing_word_kana":"ゆいいつ","containing_word_meaning":"the only/sole","type":"substring","confidence":"high"},
{"name_kana":"まこと","len":3,"romaji":"Makoto","containing_word_kana":"まことに","containing_word_meaning":"truly/indeed","type":"substring","confidence":"high"},
{"name_kana":"さくら","len":3,"romaji":"Sakura","containing_word_kana":"さくらんぼ","containing_word_meaning":"cherry (fruit)","type":"substring","confidence":"high"},
{"name_kana":"あきら","len":3,"romaji":"Akira","containing_word_kana":"あきらめる","containing_word_meaning":"to give up","type":"substring","confidence":"high"},
{"name_kana":"あすか","len":3,"romaji":"Asuka","containing_word_kana":"あすから","containing_word_meaning":"from tomorrow","type":"substring","confidence":"high"},
{"name_kana":"ひろし","len":3,"romaji":"Hiroshi","containing_word_kana":"ひろしま","containing_word_meaning":"Hiroshima","type":"substring","confidence":"high"},
{"name_kana":"ひとみ","len":3,"romaji":"Hitomi","containing_word_kana":"ひとみしり","containing_word_meaning":"shyness with strangers","type":"substring","confidence":"high"},
{"name_kana":"ちなみ","len":3,"romaji":"Chinami","containing_word_kana":"ちなみに","containing_word_meaning":"by the way/incidentally","type":"substring","confidence":"high"},
{"name_kana":"あさひ","len":3,"romaji":"Asahi","containing_word_kana":"あさひ","containing_word_meaning":"morning sun","type":"homograph","confidence":"high","grounded":"neko,hana"},
{"name_kana":"いずみ","len":3,"romaji":"Izumi","containing_word_kana":"いずみ","containing_word_meaning":"spring/fountain","type":"homograph","confidence":"high","grounded":"merosu"},
{"name_kana":"ひかる","len":3,"romaji":"Hikaru","containing_word_kana":"ひかる","containing_word_meaning":"to shine","type":"homograph","confidence":"high"},
{"name_kana":"かおり","len":3,"romaji":"Kaori","containing_word_kana":"かおり","containing_word_meaning":"fragrance/scent","type":"homograph","confidence":"high"},
{"name_kana":"","len":3,"romaji":"Nozomi","containing_word_kana":"","containing_word_meaning":"wish/hope","type":"homograph","confidence":"high"},
{"name_kana":"","len":3,"romaji":"Minami","containing_word_kana":"","containing_word_meaning":"south","type":"homograph","confidence":"high"},
{"name_kana":"","len":3,"romaji":"Hinata","containing_word_kana":"","containing_word_meaning":"a sunny spot","type":"homograph","confidence":"high"},
{"name_kana":"","len":4,"romaji":"Himawari","containing_word_kana":"","containing_word_meaning":"sunflower","type":"homograph","confidence":"high"},
{"name_kana":"","len":4,"romaji":"Nadeshiko","containing_word_kana":"","containing_word_meaning":"idealized Japanese woman/dianthus","type":"substring","confidence":"high"},
{"name_kana":"","len":4,"romaji":"Suzuran","containing_word_kana":"","containing_word_meaning":"lily of the valley","type":"homograph","confidence":"high"},
{"name_kana":"","len":4,"romaji":"Ajisai","containing_word_kana":"","containing_word_meaning":"hydrangea","type":"homograph","confidence":"high"},
{"name_kana":"","len":4,"romaji":"Tanpopo","containing_word_kana":"","containing_word_meaning":"dandelion","type":"homograph","confidence":"high"},
{"name_kana":"","len":4,"romaji":"Kisaragi","containing_word_kana":"","containing_word_meaning":"February (old lunar calendar)","type":"homograph","confidence":"med"},
{"name_kana":"","len":4,"romaji":"Minazuki","containing_word_kana":"","containing_word_meaning":"June (old lunar calendar)","type":"homograph","confidence":"med"},
{"name_kana":"","len":4,"romaji":"Kotonoha","containing_word_kana":"","containing_word_meaning":"words/language (poetic)","type":"homograph","confidence":"med"}
],
"true_positive_contexts": [
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":" ","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:merosu"},
{"name_kana":"","sentence_kana":"","source":"corpus:neko"},
{"name_kana":"","sentence_kana":"","source":"corpus:neko"},
{"name_kana":"","sentence_kana":"","source":"constructed"},
{"name_kana":"","sentence_kana":"","source":"constructed"},
{"name_kana":"","sentence_kana":"","source":"constructed"},
{"name_kana":"","sentence_kana":"","source":"constructed"},
{"name_kana":"","sentence_kana":"","source":"constructed"}
]
}