已知下图，图中给出了平均路径长度和聚类系数的数值模拟。要想生成带有最佳聚类系数的图，则最佳的边概率范围是多少？

You are given the following diagram that shows the numerical simulation of average path length and clustering coefficient. What is is the best edge probability range if you want to generate a graph with the best clustering coefficient?

Question

已知下图，图中给出了平均路径长度和聚类系数的数值模拟。要想生成带有最佳聚类系数的图，则最佳的边概率范围是多少？

You are given the following diagram that shows the numerical simulation of average path length and clustering coefficient. What is is the best edge probability range if you want to generate a graph with the best clustering coefficient?

BlackTom AI · Accepted Answer

The clustering coefficient stays essentially maximal for very small rewiring/edge probabilities, so the best range is 0.0001 – 0.001.

1. Relevant knowledge points and concepts
- The plotted curves show two normalized measures versus edge probability p: average path length L(p)/L(0) and clustering coefficient C(p)/C(0).
- In small-world models (e.g., Watts–Strogatz), small p preserves the local lattice structure and thus a high clustering coefficient; increasing p rewires edges and reduces clustering.
- The vertical axis “rate” is the fraction of the original (p=0) value; a value near 1 means clustering is nearly unchanged (high).

2. Step‑by‑step reasoning process
- Read the right-hand curve labeled C(p)/C(0). At the left-most p values (around 10^−4 to 10^−3) the clustering ratio is essentially 1 (close to the top of the plot).
- As p increases toward 10^−2 the clustering curve begins to decline noticeably, and by p ≈ 0.1 it has dropped sharply toward 0.
- Therefore the region where the clustering coefficient is best (maximal or near-original) corresponds to the smallest p interval shown on the horizontal logarithmic scale: 0.0001–0.001.

3. Intermediate/graph reading steps
- At p = 10^−4 → C(p)/C(0) ≈ 1.
- At p ≈ 10^−3 → C(p)/C(0) still ≈ 1 (no appreciable drop).
- At p ≈ 10^−2 → C(p)/C(0) begins to fall below ~0.8 and continues downward for larger p.

4. Why the correct answer is correct
- Because clustering is preserved for extremely low rewiring probabilities; only when p grows beyond ~10^−3 does rewiring significantly reduce local triangles and hence the clustering coefficient. The plotted curve confirms the maximum clustering occurs in the 0.0001–0.001 range.

5. Why other options are incorrect
- 0.001–0.01: clustering already starts to decline in this band.
- 0.01–0.1 and 0.1–1: clustering is substantially reduced (far below 1) due to heavy randomization.

6. Conclusion
- The best edge probability range for maximal clustering coefficient is 0.0001 – 0.001, matching the provided correct answer.

类似问题

在随机图 G(n, p) 中，边的数量的期望值是 m = In a random graph G(n, p), the expected number of edge m =

A project has a required return of 12.6 percent, an initial cash outflow of $42,100, and cash inflows of $16,500 in Year 1, $11,700 in Year 2, and $10,400 in Year 4. What is the net present value?

更多留学生实用工具

智能学习助手

风格化写作助手

论文查重助手

文献引用助手

课堂转译助手

课堂笔记助手

Quiz搜索助手

学校历年真题

智能刷题助手

智能匹配练习

希望你的学习变得更简单