关于「结构洞:桥接的想法从哪里来」,《结构洞、函子迁移与灵感共振》给出的定量结论是什么?

结论:第一章计算伯特的约束度:跨越结构洞使约束度从 0.629 降到 0.360,而在聚类的研究者网络中,重连 0.01 的关系就把平均距离从 17.09 步缩短到 9.98。

数据口径:书中的网络、领域、评分与势阱均为示意模型;全部计算由书中程序生成;本书不是投资、医疗或心理建议。 第六章用模拟检验克莱默斯公式(相差不超过 1.32 倍),并发现噪声太少是主要风险:低噪声下只有 0.017 的运行找到更好的状态,而先孵化后专注的节奏(0.730)并不比稳定的中等噪声(0.750)更好。

English version

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: 0.01, 17.09, 9.98

rewiring pclusteringmean pathlong ties
0.000.60017.090
0.010.5849.986
0.050.5195.2430
0.200.3203.76120
1.000.0243.15600
查看完整验算程序
# Chapter 1. p02: How few bridges it takes. A ring of 200 researchers,
# each tied to the 3 nearest on each side, has high clustering and long
# paths. Rewire each tie with chance p to a random researcher (Watts and
# Strogatz). We print clustering and mean shortest path; a handful of
# long ties shortens paths sharply while clustering barely moves.
# Fixed seed 102; averages over 10 networks.
import math
import random
from collections import deque

rng = random.Random(102)
N, K = 200, 3

def ring(p):
    edges = set()
    for i in range(N):
        for d in range(1, K + 1):
            a, b = i, (i + d) % N
            if rng.random() < p:
                while True:
                    b = rng.randrange(N)
                    e = (min(a, b), max(a, b))
                    if b != a and e not in edges:
                        break
            edges.add((min(a, b), max(a, b)))
    g = [set() for _ in range(N)]
    for a, b in edges:
        g[a].add(b)
        g[b].add(a)
    return g

def clustering(g):
    cs = []
    for nb in g:
        k = len(nb)
        if k < 2:
            continue
        links = sum(1 for u in nb for v in nb if u < v and v in g[u])
        cs.append(links / (k * (k - 1) / 2))
    return math.fsum(cs) / len(cs)

def mean_path(g):
    tot = cnt = 0
    for s in range(N):
        dist = {s: 0}
        q = deque([s])
        while q:
            u = q.popleft()
            for v in g[u]:
                if v not in dist:
                    dist[v] = dist[u] + 1
                    q.append(v)
        tot += sum(dist.values())
        cnt += len(dist) - 1
    return tot / cnt

if __name__ == "__main__":
    print("   rewiring p   clustering   mean path   long ties")
    for p in (0.0, 0.01, 0.05, 0.2, 1.0):
        c = m = 0.0
        for _ in range(10):
            g = ring(p)
            c += clustering(g) / 10
            m += mean_path(g) / 10
        print("   {:10.2f}   {:10.3f}   {:9.2f}   {:9.0f}".format(
            p, c, m, p * N * K))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《跨学科结构洞桥接、函子知识迁移与灵感随机共振》第 1 章「结构洞:桥接的想法从哪里来」(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-102 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026102,
  author    = {Fan, Yafu},
  title     = {Cross-Disciplinary Structural Holes, Functorial Knowledge Transfer and Inspiration Stochastic Resonance},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论