{
  "version": "1.0",
  "source_chunks_path": "01_chunks.json",
  "strategy": {
    "front_ratio": 0.15,
    "middle_positions": [
      0.25,
      0.5,
      0.75
    ],
    "tail_ratio": 0.1,
    "high_density_limit": 30,
    "front_cap": 12,
    "tail_cap": 8,
    "middle_radius": 1,
    "preview_chars": 220
  },
  "samples": [
    {
      "sample_id": "OS0001",
      "chunk_id": "C0001",
      "start_para": 1,
      "end_para": 2,
      "sampling_reason": "front_loaded",
      "char_len": 1786,
      "text_preview": "第一回 楔子 上海地方，為商賈麇集之區，中外雜處，人煙稠密，輪舶往來，百貨輸轉。加以蘇揚各地之煙花，亦都圖上海富商大賈之多，一時買棹而來，環聚於四馬路一帶，高張豔幟，炫異爭奇。那上等的，自有那一班王孫公子去問津；那下等的，也有那些逐臭之夫，垂涎著要嘗鼎一臠。於是乎把六十年前的一片蘆葦灘頭，變做了中國第一個熱鬧的所在。唉！繁華到極，便容易淪於虛浮。久而久之，凡在上海來來往往的人，開口便講應酬，閉口也講應酬。人生世上，這「應酬」兩個字，本來...",
      "chunk_index": 1,
      "all_reasons": [
        "front_loaded",
        "high_entity_density"
      ],
      "density_score": 5.25,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "應酬",
          "空心大老官",
          "死裡逃生",
          "這書要賣也可以，要不賣也可以",
          "不賣呢"
        ],
        "named_places": [
          "到了上海",
          "然後出城"
        ],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [
          "到了上海",
          "然後出城"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 5,
          "named_places": 2,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 2
        }
      }
    },
    {
      "sample_id": "OS0002",
      "chunk_id": "C0002",
      "start_para": 2,
      "end_para": 3,
      "sampling_reason": "front_loaded",
      "char_len": 157,
      "text_preview": "想定了主意，就將這冊子的記載，改做了小說體裁，剖作若干回，加了些評語，寫一封信，另外將冊子封好，寫著「寄日本橫濱市山下町百六十番新小說社」。走到虹口蓬路日本郵便局，買了郵稅票黏上，交代明白，翻身就走。一直走到深山窮谷之中，絕無人煙之地，與木石居，與鹿豕游去了。 第二回 守常經不使疏逾戚 睹怪狀幾疑賊是官",
      "chunk_index": 2,
      "all_reasons": [
        "front_loaded",
        "high_entity_density"
      ],
      "density_score": 0.039,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 0,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0003",
      "chunk_id": "C0003",
      "start_para": 3,
      "end_para": 4,
      "sampling_reason": "middle_stratified",
      "char_len": 1781,
      "text_preview": "第二回 守常經不使疏逾戚 睹怪狀幾疑賊是官 新小說社記者接到了死裡逃生的手書及九死一生的筆記，展開看了一遍，不忍埋沒了他，就將他逐期刊布出來。閱者須知，自此以後之文，便是九死一生的手筆，及死裡逃生的批評了。我是好好的一個人，生平並未遭過大風波、大險阻，又沒有人出十萬兩銀子的賞格來捉我，何以將自己好好的姓名來隱了，另外叫個甚麼九死一生呢？只因我出來應世的二十年中，回頭想來，所遇見的只有三種東西：第一種是蛇蟲鼠蟻；第二種是豺狼虎豹；第三種是...",
      "chunk_index": 3,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 5.5,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "是，我父親同他是相好",
          "世伯何以知道他靠不住呢",
          "今日張鼎臣同你說些甚麼",
          "並未說甚麼。他問我討主意，我說沒有主意",
          "應該寄多少呢",
          "自然是愈多愈好呀"
        ],
        "named_places": [
          "先到上海",
          "方到杭州"
        ],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [
          "先到上海",
          "方到杭州"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 6,
          "named_places": 2,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 2
        }
      }
    },
    {
      "sample_id": "OS0004",
      "chunk_id": "C0004",
      "start_para": 4,
      "end_para": 4,
      "sampling_reason": "middle_stratified",
      "char_len": 1783,
      "text_preview": "我伯父看見了，便立起來問道：「這訃帖底稿，是哪個起的呢？」我說道：「就是姪兒起的。」我的伯父拿起來一看，對著張鼎臣說道：「這才是吾家千里駒呢。這訃聞居然是大大方方的，期、功、緦麻，一點也沒有弄錯。」鼎臣看著我，笑了一笑，並不回言。伯父又指著訃帖當中一句問我道：「你父親今年四十五歲，自然應該作『享壽四十五歲』，為甚你卻寫做『春秋四十五歲』呢？」我說道：「四十五歲，只怕不便寫作『享壽』。有人用的是『享年』兩個字。姪兒想去，年是說不著享的；若...",
      "chunk_index": 4,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 5.5,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "享年",
          "得年",
          "存年",
          "春秋",
          "這小小年紀，難得他這等留心呢",
          "家母年紀又不很大，哪裡會善忘到這麼著",
          "捉賊捉贓呀，你捉著贓沒有呢",
          "王八"
        ],
        "named_places": [
          "先到上海",
          "還留在杭州"
        ],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [
          "先到上海",
          "還留在杭州"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 10,
          "named_places": 2,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 2
        }
      }
    },
    {
      "sample_id": "OS0005",
      "chunk_id": "C0005",
      "start_para": 4,
      "end_para": 5,
      "sampling_reason": "middle_stratified",
      "char_len": 1662,
      "text_preview": "還有兩個人，都穿的是藍布長衫，像是個底下人光景。我想這明明是個官場中人，如何會做賊呢？這廣東人太胡鬧了。只聽那廣東人又對眾人說道：「我不說明白，你們眾人一定說我錯疑了人了；且等我說出來，大眾聽聽呀。我父子兩人同來。我住的房艙，是在外南，房門口對著江面的。我們已經睡了，忽聽得我兒子叫了一聲：『有賊！』我一咕嚕爬進來看時，兩件熟羅長衫沒了；衣箱面上擺的一個小鬧鐘，也不見了；衣箱的鎖，也幾乎撬開了。我便追出來，轉個彎要進裡面，便見這個人在當路...",
      "chunk_index": 5,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 3.9,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "有賊",
          "當路站著，如何便可說他做賊呢",
          "晚上睡不著，出去望望也是常事。怎麼便說他望風",
          "你讓我搜麼",
          "我只問你要",
          "你要東西跟我來",
          "東西在這個裡面"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [
          "有分教"
        ],
        "artifacts": [],
        "term_highlights": [
          "有分教"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 7,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 1,
          "artifacts": 0,
          "conservative_terms": 1
        }
      }
    },
    {
      "sample_id": "OS0006",
      "chunk_id": "C0006",
      "start_para": 5,
      "end_para": 6,
      "sampling_reason": "middle_stratified",
      "char_len": 1720,
      "text_preview": "第三回 走窮途忽遇良朋 談仁路初聞怪狀 卻說我搬到客棧裡住了兩天，然後到伯父公館裡去打聽，說還沒有回來。我只得耐心再等。一連打聽了幾次，卻只不見回來。我要請見伯母，他又不肯見，此時我已經住了十多天，帶來的盤纏，本來沒有多少，此時看看要用完了，心焦的了不得。這一天我又去打聽了，失望回來，在路上一面走，一面盤算著：倘是過幾天還不回來，我這裡莫說回家的盤纏沒有，就是客棧的房飯錢，也還不曉得在那裡呢！正在那裡納悶，忽聽得一個人提著我的名字叫我。...",
      "chunk_index": 6,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 1.65,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "是個同知班",
          "有欠過房飯錢麼",
          "承大哥過愛，下榻在此，理當要請見大嫂才是",
          "說來話長呢。你先要懂得『野雞",
          "土老兒"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 5,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0007",
      "chunk_id": "C0007",
      "start_para": 6,
      "end_para": 6,
      "sampling_reason": "middle_stratified",
      "char_len": 1409,
      "text_preview": "繼之道：「跑街是到外面收帳的意思。有時到外面打聽行情，送送單子，也是他的事。這土老兒做了一年多，倒還安分。一天不知聽了甚麼人說起『打野雞』的好處，……」我聽了，又不明白道：「甚麼打野雞？可是打那流娼麼？」繼之道：「去嫖流娼，就叫打野雞。這土老兒聽得心動，那一天帶了幾塊洋錢，走到了四馬路野雞最多的地方，叫做甚麼會香裡，在一家門首，看見一個『黃魚』。」我聽了，又是一呆道：「甚麼叫做黃魚？」繼之道：「這是我說錯南京的土談了，這裡南京人，叫大腳...",
      "chunk_index": 7,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 1.852,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "打野雞",
          "甚麼打野雞？可是打那流娼麼",
          "黃魚",
          "甚麼叫做黃魚",
          "明天來",
          "乾濕",
          "裝乾濕",
          "六塊洋錢"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 18,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0008",
      "chunk_id": "C0008",
      "start_para": 6,
      "end_para": 6,
      "sampling_reason": "middle_stratified",
      "char_len": 1149,
      "text_preview": "「到了次日，桂花叫土老兒去錢莊裡辭了職役。土老兒果然依了他的話。但回頭一想，恐怕這件事不妥當，到後來要再謀這麼一件事就難了。於是打了一個主意，去見東家，先撒一個謊說：『家裡有要緊事，要請個假回去一趟，頂多兩三個月就來的。』東家准了。這是他的意思，萬一不妥當，還想後來好回去仍就這件事。於是取了鋪蓋，直跑到會香裡，同桂花住了幾天。桂花帶了土老兒到京城裡去，居然同他捐了一個二品頂戴的道臺，還捐了一枝花翎，辦了引見，指省江蘇。在京的時候，土老兒...",
      "chunk_index": 8,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 0.287,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 0,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0009",
      "chunk_id": "C0009",
      "start_para": 6,
      "end_para": 7,
      "sampling_reason": "middle_stratified",
      "char_len": 723,
      "text_preview": "繼之道：「這是前兩年的事了。前兩年制臺得了個心神彷彿的病。年輕時候，本來是好色的；到如今偌大年紀，他那十七八歲的姨太太，還有六七房，那通房的丫頭，還不在內呢。他這好色的名出了，就有人想拿這個巴結他。他病了的時候，有一個年輕的候補道，自己陳說懂得醫道。制臺就叫他診脈。他診了半晌說：『大帥這個病，卑職不能醫，不敢胡亂開方；卑職內人怕可以醫得。』制臺道：『原來尊夫人懂得醫理，明日就請來看看罷。』到了明日，他的那位夫人，打扮得花枝招展的來了。診...",
      "chunk_index": 9,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 1.431,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "原來尊夫人懂得醫理，明日就請來看看罷",
          "這個病不必吃藥，只用按摩之法，就可以痊癒",
          "妾頗懂得",
          "哼，你說他沒有臉住這裡麼？他還得意得很呢",
          "這還有甚麼得意之處呢"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 5,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0010",
      "chunk_id": "C0010",
      "start_para": 7,
      "end_para": 8,
      "sampling_reason": "middle_stratified",
      "char_len": 1660,
      "text_preview": "第四回 吳繼之正言規好友 苟觀察致敬送嘉賓 卻說我追問繼之：「那一個候補道，他的夫人受了這場大辱，還有甚麼得意？」繼之道：「得意呢！不到十來天工夫，他便接連著奉了兩個札子，委了籌防局的提調以及山貨局的會辦了。去年還同他開上一個保舉。他本來只是個鹽運司銜，這一個保舉，他就得了個二品頂戴了。你說不是得意了嗎？」我聽了此話，不覺呆了一呆道：「那麼說，那一位總督大帥，竟是被那一位夫人……」我說到此處，以下還沒有說出來，繼之便搶著說道：「那個且不...",
      "chunk_index": 10,
      "all_reasons": [
        "middle_stratified",
        "high_entity_density"
      ],
      "density_score": 0.9,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "奇怪，奇怪",
          "沒有"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 2,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0011",
      "chunk_id": "C0011",
      "start_para": 8,
      "end_para": 8,
      "sampling_reason": "tail_stratified",
      "char_len": 1785,
      "text_preview": "繼之又道：「雖是這麼說，你也不必著急。我今天見了藩臺，他說此地大關的差使，前任委員已經滿了期了，打算要叫我接辦，大約一兩天就可以下札子。我那裡左右要請朋友，你就可以揀一個合式的事情，代我辦辦。我們是同窗至好，我自然要好好的招呼你。至於你令伯的話，只好慢慢再說，好在他終久是要回來的，總不能一輩子不見面。」我說道：「家伯到通州去的話，可是大哥打聽來的，還是別人傳說的呢？」繼之道：「這是我在藩署號房打聽來的，千真萬真，斷不是謠言。你且坐坐，我...",
      "chunk_index": 11,
      "all_reasons": [
        "tail_stratified",
        "high_entity_density"
      ],
      "density_score": 0.9,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "沒有",
          "到那裡去過"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 2,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    },
    {
      "sample_id": "OS0012",
      "chunk_id": "C0012",
      "start_para": 8,
      "end_para": 8,
      "sampling_reason": "tail_stratified",
      "char_len": 427,
      "text_preview": "再看那主人時，卻放下了馬蹄袖，拱起雙手，一直拱到眉毛上面，彎著腰，嘴裡不住的說：「請，請，請！」直到那客人走的轉了個彎看不見了，方才進去，「呀」的一聲，大門關了。我再留心看那門口時，卻掛著一個紅底黑字的牌兒，像是個店家招牌。再看看那牌上的字，卻寫的是「欽命二品頂戴，賞戴花翎，江蘇即補道，長白苟公館」二十個宋體字。不覺心中暗暗納罕。走到前面，僱定了馬匹，騎到關上去，見過繼之。這天沒有甚麼事，大家坐著閒談一會。開出午飯來，便有幾個同事都過來...",
      "chunk_index": 12,
      "all_reasons": [
        "tail_stratified",
        "high_entity_density"
      ],
      "density_score": 0.357,
      "signal_terms": {
        "chapter_headings": [],
        "quoted_phrases": [
          "請，請，請"
        ],
        "named_places": [],
        "titled_people": [],
        "organizations": [],
        "artifacts": [],
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 1,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    }
  ],
  "stats": {
    "chunk_count": 12,
    "sample_count": 12,
    "counts_by_primary_reason": {
      "front_loaded": 2,
      "middle_stratified": 8,
      "tail_stratified": 2
    },
    "counts_by_any_reason": {
      "front_loaded": 2,
      "high_entity_density": 12,
      "middle_stratified": 8,
      "tail_stratified": 2
    },
    "top_density_chunks": [
      {
        "chunk_id": "C0003",
        "chunk_index": 3,
        "density_score": 5.5,
        "term_highlights": [
          "先到上海",
          "方到杭州"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 6,
          "named_places": 2,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 2
        }
      },
      {
        "chunk_id": "C0004",
        "chunk_index": 4,
        "density_score": 5.5,
        "term_highlights": [
          "先到上海",
          "還留在杭州"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 10,
          "named_places": 2,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 2
        }
      },
      {
        "chunk_id": "C0001",
        "chunk_index": 1,
        "density_score": 5.25,
        "term_highlights": [
          "到了上海",
          "然後出城"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 5,
          "named_places": 2,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 2
        }
      },
      {
        "chunk_id": "C0005",
        "chunk_index": 5,
        "density_score": 3.9,
        "term_highlights": [
          "有分教"
        ],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 7,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 1,
          "artifacts": 0,
          "conservative_terms": 1
        }
      },
      {
        "chunk_id": "C0007",
        "chunk_index": 7,
        "density_score": 1.852,
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 18,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      },
      {
        "chunk_id": "C0006",
        "chunk_index": 6,
        "density_score": 1.65,
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 5,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      },
      {
        "chunk_id": "C0009",
        "chunk_index": 9,
        "density_score": 1.431,
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 5,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      },
      {
        "chunk_id": "C0010",
        "chunk_index": 10,
        "density_score": 0.9,
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 2,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      },
      {
        "chunk_id": "C0011",
        "chunk_index": 11,
        "density_score": 0.9,
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 2,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      },
      {
        "chunk_id": "C0012",
        "chunk_index": 12,
        "density_score": 0.357,
        "term_highlights": [],
        "signal_counts": {
          "chapter_headings": 0,
          "quoted_phrases": 1,
          "named_places": 0,
          "titled_people": 0,
          "organizations": 0,
          "artifacts": 0,
          "conservative_terms": 0
        }
      }
    ]
  },
  "resolved_strategy": {
    "front_count": 2,
    "tail_count": 2,
    "middle_radius": 1,
    "preview_chars": 220
  }
}