关于「间隔指纹:突发性系数、记忆系数与「量少就安全」的错觉」,《公域反垃圾检测与合规桥接页》给出的定量结论是什么?
结论:第一章说明突发性系数与发帖频率无关:固定定时器为 −0.969,类人节奏为 0.300。
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分析,不构成经营、法律或理财建议,法规与平台规则以上架前人工复核为准。 书中以标注框区分已发表的证据、可运行的 Python 计算(20 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: 0.300
| rhythm | mean gap h | B | M | H bits |
|---|---|---|---|---|
| fixed scheduler | 6.00 | -0.969 | 0.025 | 0.99 |
| loose scheduler | 5.85 | -0.445 | 0.009 | 4.99 |
| Poisson | 3.81 | -0.005 | 0.001 | 5.23 |
| human-like | 5.28 | 0.300 | -0.003 | 4.60 |
| reference texts | B 0.00 | M 0.01 | (Goh and Barabasi 2008) | |
| reference heart | B -0.69 | M 0.97 | (healthy, same paper) | |
| periodic limit B -1; Poisson B 0; most bursty limit B 1 |
查看完整验算程序
# Chapter 1. p11: burstiness B and memory M of four posting rhythms.
# Model: Goh and Barabasi (EPL 2008) define B = (s - m) / (s + m) on the
# inter-event times (m mean, s standard deviation) and M as the
# correlation of consecutive inter-event times. Four synthetic rhythms,
# 400 gaps each, mean gap about 4 to 6 hours (illustrative):
# fixed scheduler 6 h with uniform jitter of +-10 minutes
# loose scheduler uniform 2 to 10 hours
# Poisson exponential, mean 4 h
# human-like session mixture: 0.6 short gaps (exp, mean 0.25 h)
# and 0.4 long gaps (lognormal, median 8 h, sigma 1.0)
# H is the Shannon entropy (bits) of the gaps rounded to 15 minutes,
# the timing-entropy idea of Chu et al. (IEEE TDSC 2012).
# Published reference points (Goh and Barabasi 2008): letter spacing in
# texts B 0, M 0.01; healthy heartbeat B -0.69, M 0.97.
import math
import random
import statistics
from collections import Counter
def burstiness(gaps):
m = statistics.fmean(gaps)
s = statistics.pstdev(gaps)
return (s - m) / (s + m)
def memory(gaps):
return statistics.correlation(gaps[:-1], gaps[1:])
def entropy_bits(gaps, step=0.25):
c = Counter(round(g / step) for g in gaps)
n = len(gaps)
return -math.fsum((k / n) * math.log2(k / n) for k in c.values())
def human_gaps(rng, n):
out = []
for _ in range(n):
if rng.random() < 0.6:
out.append(rng.expovariate(1 / 0.25))
else:
out.append(rng.lognormvariate(math.log(8.0), 1.0))
return out
def rhythms(rng, n):
return [
("fixed scheduler", [6 + rng.uniform(-1 / 6, 1 / 6)
for _ in range(n)]),
("loose scheduler", [rng.uniform(2, 10) for _ in range(n)]),
("Poisson", [rng.expovariate(1 / 4) for _ in range(n)]),
("human-like", human_gaps(rng, n)),
]
if __name__ == "__main__":
rng = random.Random(20261120)
print("rhythm mean gap h B M H bits")
for name, g in rhythms(rng, 400):
print("{:<16} {:>10.2f} {:>7.3f} {:>7.3f} {:>8.2f}".format(
name, statistics.fmean(g), burstiness(g), memory(g),
entropy_bits(g)))
print("reference texts B 0.00 M 0.01 (Goh and Barabasi 2008)")
print("reference heart B -0.69 M 0.97 (healthy, same paper)")
print("periodic limit B -1; Poisson B 0; most bursty limit B 1")唯一数据源与引用
范雅福(Fan, Y.). (2026). 《公域社交图谱反垃圾检测边界、合规联盟桥接页转化与限流自愈动力学母典》第 1 章「间隔指纹:突发性系数、记忆系数与「量少就安全」的错觉」(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-342 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026342,
author = {Fan, Yafu},
title = {Social-Graph Anti-Spam Detection Bounds, Compliant Affiliate Bridge-Page Funnels and the Kinetics of Recovery after Reach Throttling},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}