关于「遗忘与复习:半衰期回归与一份诚实的时间表」,《词汇覆盖率与速成边界》给出的定量结论是什么?
结论:第五章用半衰期回归排复习表:「九十天速成」只能走到 0.8050 的覆盖率,每天约半小时要 500 天才到八千词族。
数据口径:书中参数为示意,全部计算由书中程序生成;本书是非个人化的经营与教学参考,不是针对任何个人或企业的经营建议,也不承诺任何课程能在某个期限内带来流利。 书中以标注框区分已发表的证据、可运行的 Python 计算(19 段程序)、原典与作者的假说。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.8050, 500
| new/day | min/day | days to 2000 | to 4500 | to 8000 | cov 90 d | cov 365 d |
| 10 | 15.3 | 250 | 563 | 1000 | 0.6445 | 0.9114 |
| 20 | 30.7 | 125 | 282 | 500 | 0.8050 | 0.9661 |
| 40 | 61.3 | 63 | 141 | 250 | 0.9099 | 0.9911 |
| a '90-day' plan at 20 a day: 1440 families, coverage 0.8050, | ||||||
| i.e. one unknown word in 5.1 running words |
查看完整验算程序
# Chapter 5. p53: an honest timeline from words a day to coverage
# A plan introduces N new families a day; a share G = 0.8 of them
# (illustrative) end up known for reading. Known families after t days
# are G N t. Coverage is read off the reference family curve of p23
# (V = 20000, s = 2.389, q = 664.5); daily minutes come from the day-180
# load of p52 (8 s a review, 20 s a new item, about 9 reviews a word).
# Targets: 2000 families, 4500 (95 % anchor) and 8000 (98 % anchor).
import math
V, S, Q, G = 20000, 2.389, 664.5, 0.8
REVIEWS, SEC_REV, SEC_NEW = 9, 8, 20
def partial(n, s, q):
return math.fsum([(r + q) ** (-s) for r in range(1, n + 1)])
TOTAL = partial(V, S, Q)
def cover(n):
return partial(min(int(n), V), S, Q) / TOTAL
if __name__ == "__main__":
print("new/day min/day days to 2000 to 4500 to 8000"
" cov 90 d cov 365 d")
for n in (10, 20, 40):
mins = (n * REVIEWS * SEC_REV + n * SEC_NEW) / 60.0
days = [math.ceil(t / (G * n)) for t in (2000, 4500, 8000)]
print(" {:5d} {:7.1f} {:12d} {:7d} {:7d} {:8.4f} {:9.4f}"
.format(n, mins, days[0], days[1], days[2],
cover(G * n * 90), cover(G * n * 365)))
known = G * 20 * 90
print("a '90-day' plan at 20 a day: {:.0f} families, coverage {:.4f},"
.format(known, cover(known)))
print(" i.e. one unknown word in {:.1f} running words".format(
1.0 / (1.0 - cover(known))))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《二语词汇覆盖率九十八阈值、齐普夫频率曲线与马来语闽南语速成边界母典》第 5 章「遗忘与复习:半衰期回归与一份诚实的时间表」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-228 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026228,
author = {Fan, Yafu},
title = {The 98% Lexical-Coverage Threshold, Zipfian Frequency Curves and Fast-Track Limits for Malay and Hokkien Learners},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}