主成分分析（PCA）在做什么样的数学？
What mathematics does Principal Component Analysis (PCA) do?
它在高维数据中找方差最大的方向（特征向量），把数据投影到低维而尽量保留信息，本质是对协方差矩阵做特征值分解。是降维的核心工具。
It finds directions of maximal variance (eigenvectors) in high-dimensional data and projects onto them to keep information while reducing dimension—essentially eigendecomposition of the covariance matrix. A core dimensionality-reduction tool.
难度 专业 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-10-04
来源 · Source
H. Hotelling《Analysis of a Complex Statistical Variable into Principal Components》, Journal of Educational Psychology 24(6):417–441, 1933
H. Hotelling, 'Analysis of a Complex Statistical Variable into Principal Components', Journal of Educational Psychology 24(6):417–441, 1933
独立核验依据 · Independent verification: 核验依据 1 DOI 10.1007/978-3-540-72244-1_9
#建模 #降维
随机过程里的‘泊松过程’在描述什么？
What does the Poisson process describe in stochastic processes?
它刻画独立、平均速率恒定的随机事件到达，如来电、放射性衰变。在时长 t 内发生 k 次的概率呈泊松分布，间隔则是指数分布。
It models independent random events arriving at a constant average rate, like calls or radioactive decays. The count in time t follows a Poisson distribution; intervals are exponential.
难度 专业 · 适合 高中(15-18)
CC BY 4.0 · 已核验 · Verified 2026-11-02
来源 · Source
美国数学会(AMS)《随机过程导论》
American Mathematical Society (AMS), Introduction to Stochastic Processes
独立核验依据 · Independent verification: 核验依据 1 ISBN 9787111315445（豆瓣书目条目: 随机过程导论）
#随机 #泊松
单纯形法是用来做什么的？
What is the simplex method used for?
它是求解线性规划（在线性约束下最优化线性目标）的经典算法，沿多面体顶点一步步走到最优解。虽最坏情况是指数时间，但实践中通常很快。
It is the classic algorithm for linear programming (optimizing a linear objective under linear constraints), walking vertex to vertex of a polytope to the optimum. Though exponential in worst case, it is usually fast in practice.
难度 专业 · 适合 大众
CC BY 4.0 · 来源待核验 · Source pending review
来源 · Source
出处备注：著录仅给书名，无版次与年份；同一书名常对应多个版次，不臆测具体版本。
《运筹学》Hillier 与 Lieberman
Introduction to Operations Research by Hillier and Lieberman
#优化 #线性规划
在没有出生与死亡的封闭人群中，经典 SIR 模型为什么预言疫情会在仍有许多易感者未感染时就自行终止？终态规模方程揭示了什么规律？
In a closed population with no births or deaths, why does the classical SIR model predict that an epidemic dies out while many susceptibles remain uninfected, and what does the final size equation reveal?
终态规模方程 ln(s∞/s₀)=−R₀(s₀−s∞)（s₀≈1 时即 ln s∞=−R₀(1−s∞)）把结束时剩余易感比例与 R₀ 联系起来；R₀>1 时除平凡根外仅有一支 s∞<1 的正根。传播停止并非因易感者耗尽，而是易感比例降至约 1/R₀、有效再生数跌破 1 后感染链逐渐衰减，故总会留下从未被感染的人群；由 R₀ 与初值即可唯一确定总感染规模。
The final size relation, ln(s∞/s0) = −R0(s0 − s∞), which reduces to ln s∞ = −R0(1 − s∞) when s0 ≈ 1, ties the fraction of people still susceptible when transmission fades to the basic reproduction number; for R0 > 1 it has, besides the trivial root, a unique solution below 1. An epidemic stops not because susceptibles run out, but because once susceptibility falls near 1/R0 the effective reproduction number drops below one and infection chains decay. Many individuals therefore remain never infected, and the attack rate is pinned down by R0 and the initial conditions alone.
难度 专业 · 适合 高中(15-18)
CC BY 4.0 · 已核验 · Verified 2026-11-03
来源 · Source
O. Diekmann、J.A.P. Heesterbeek《传染病数学流行病学：模型构建、分析与解释》，John Wiley & Sons，2000 年
O. Diekmann & J.A.P. Heesterbeek, Mathematical Epidemiology of Infectious Diseases: Model Building, Analysis and Interpretation, John Wiley & Sons, 2000
独立核验依据 · Independent verification: 核验依据 1 ARCHIVE 标识 mathematicalepid0000diek：Mathematical epidemi
#流行病数学模型 #终态规模 #基本再生数 #SIR模型
钟形曲线（正态分布）为什么到处出现？
Why does the bell curve (normal distribution) appear everywhere?
中心极限定理说：大量独立微小的随机误差相加，其和的分布趋近正态，与个别误差形状无关。所以测量误差、身高、考试分常呈钟形。
The central limit theorem says sums of many independent tiny random errors tend to a normal distribution, regardless of individual shapes. Hence measurement errors, heights and exam scores often look bell-shaped.
难度 进阶 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-10-04 · 3 条独立证据
来源 · Source
Feller,《An Introduction to Probability Theory and Its Applications》第3版,Wiley,1968
W. Feller, An Introduction to Probability Theory and Its Applications, 3rd ed., Wiley, 1968
独立核验依据 · Independent verification: 核验依据 1 ISBN 9789994311071 (1968) · 核验依据 2 ISBN 9780821807491 (1997) · 核验依据 3 ISBN 9789994311071 (1991)
#统计 #中心极限
指数增长翻一倍的时间固定，这有什么反直觉的后果？
What counterintuitive consequence follows from a fixed doubling time?
因为翻倍是乘法而非加法，看似‘还很早’的阶段其实已走过一半路程；等你觉得‘突然很多’时，往往只剩一次翻倍的时间了。这常被用来警示资源与污染。
Since doubling is multiplicative, the 'still early' stage is already halfway; by the time it looks 'suddenly huge', often only one more doubling remains. It warns about resources and pollution.
难度 进阶 · 适合 初中(12-15)
CC BY 4.0 · 已核验 · Verified 2026-10-04
来源 · Source
《非线性动力学与混沌》Strogatz
Nonlinear Dynamics and Chaos by Steven Strogatz
独立核验依据 · Independent verification: 核验依据 1 ISBN 9780813349107 (2024)
#建模 #指数
为什么‘加权平均’比‘简单平均’更公平？
Why is a 'weighted average' fairer than a 'simple average'?
不同数据的重要性或可靠性不同：考试里期末比平时权重高，调查里大群体样本权重更大。加权让结果反映真实贡献而非机械平分。
Data differ in importance or reliability: finals weigh more than homework, larger groups weigh more in surveys. Weighting reflects true contribution instead of mechanical equality.
难度 进阶 · 适合 初中(12-15)
CC BY 4.0 · 来源待核验 · Source pending review
来源 · Source
出处备注：本题原挂出处与答案内容不符，已撤回；具体书目待补，暂不展示来源。
本题原挂出处与答案内容不符，已撤回；具体书目待补。
Previous citation withdrawn as unsupported for this answer; a specific source is pending.
#统计 #加权
‘六度分隔’说人与人之间很短，这有数学根据吗？
Does 'six degrees of separation' have a mathematical basis?
在大规模随机或幂律网络中，路径长度随节点数对数增长，因此全球几十亿人之间平均只需很少几步。小世界网络模型解释了这种现象。
In large random or scale-free networks, path length grows only logarithmically with nodes, so billions of people are only a few steps apart on average. Small-world network models explain it.
难度 进阶 · 适合 小学高段(9-12)
CC BY 4.0 · 已核验 · Verified 2026-11-04
来源 · Source
Barabási 与 Albert,《Emergence of Scaling in Random Networks》,Science,1999 (DOI: 10.1126/science.286.5439.509)
Barabási, A.-L. & Albert, R., 'Emergence of Scaling in Random Networks', Science, 1999, DOI: 10.1126/science.286.5439.509
独立核验依据 · Independent verification: 核验依据 1 Crossref DOI 10.1126/science.286.5439.509
#网络 #小世界
旅行商问题（TSP）为什么难？
Why is the Traveling Salesman Problem (TSP) hard?
要在 n 个城市间找最短回路，候选路线随 n 阶乘式增长，没有已知的高效精确算法。它是经典的 NP 难组合优化问题，实际常用近似或启发式求解。
Finding the shortest tour through n cities has factorial-scale candidates and no known efficient exact algorithm. It is a classic NP-hard combinatorial optimization problem, usually solved by approximation or heuristics.
难度 进阶 · 适合 小学低段(6-8)
CC BY 4.0 · 已核验 · Verified 2026-10-04
来源 · Source
Applegate、Bixby、Chvatal 与 Cook,《The Traveling Salesman Problem》,Princeton University Press,2006
D. L. Applegate et al., The Traveling Salesman Problem, Princeton University Press, 2006
独立核验依据 · Independent verification: 核验依据 1 ISBN 9780691129938 (2006)
#建模 #组合优化
为什么‘相关’不等于‘因果’是建模的铁律？
Why is 'correlation is not causation' a hard rule in modeling?
两件事同时变化可能只因共同受第三因素驱动（如冰淇淋销量与溺水都随气温升）。把相关当因果会得出错误干预，建模必须谨慎区分。
Two things moving together may share a third driver (ice-cream sales and drownings both rise with temperature). Treating correlation as cause yields wrong interventions; modeling must distinguish them.
难度 进阶 · 适合 小学高段(9-12)
CC BY 4.0 · 已核验 · Verified 2026-10-04
来源 · Source
Pearl、Glymour 与 Jewell《Causal Inference in Statistics: A Primer》（Wiley）
Pearl, Glymour and Jewell, Causal Inference in Statistics: A Primer, Wiley
独立核验依据 · Independent verification: 核验依据 1 ISBN 978-1-119-18687-8 (2016, Wiley)
#建模 #因果
某地区人口按Logistic模型演化：dP/dt = 0.04·P·(1 − P/500)，其中P以万为单位，t以年为单位。若当前人口为100万，约需多少年达到250万？并解释建模中r与K的现实含义。
A region's population follows the logistic equation dP/dt = 0.04·P·(1 − P/500), with P measured in ten-thousands and t in years. If the current population is one million, how long until it reaches 2.5 million, and what real quantities do the parameters r and K represent?
方程的解为P(t)=K/(1+Ae^(−rt))，其中A=(K−P₀)/P₀=(500−100)/100=4。代入P=250得1+4e^(−0.04t)=2，解出t=ln4/0.04≈34.7，即约35年后达到250万。这里r=0.04是资源不受限制时的内禀人均增长率，K=500万是由食物、住房与环境承载能力决定的容纳量，且人口恰在K/2处增长最快。
The solution is P(t) = K/(1 + Ae^(−rt)) with A = (K − P₀)/P₀ = (500 − 100)/100 = 4. Substituting P = 250 gives 1 + 4e^(−0.04t) = 2, so t = ln4/0.04 ≈ 34.7, roughly 35 years. Here r = 0.04 is the intrinsic per-capita growth rate under unlimited resources, while K = 5 million is the carrying capacity set by food, housing and environmental limits; growth is fastest exactly at K/2.
难度 进阶 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-11-02
来源 · Source
Pearl, R. 与 Reed, L. J.《On the Rate of Growth of the Population of the United States since 1790 and Its Mathematical Representation》, PNAS, 1920年第6卷第6期, 275-288页
R. Pearl and L. J. Reed, 'On the Rate of Growth of the Population of the United States since 1790 and its Mathematical Representation', Proceedings of the National Academy of Sciences, vol. 6, no. 6, pp. 275-288, 1920.
独立核验依据 · Independent verification: 核验依据 1 CROSSREF 标识 10.1073/pnas.6.6.275：On the Rate of Growth o
#数学建模 #Logistic人口模型
如何用一组微分方程刻画传染病在人群中的传播,也就是 SIR 模型?基本再生数 R0 怎样决定疫情的走向?
How can a system of differential equations capture the spread of an infectious disease through the SIR model, and how does the basic reproduction number R0 govern the course of an epidemic?
SIR 模型把人群分成易感者 S、感染者 I 与移出者 R 三类,满足 dS/dt=-bSI/N、dI/dt=bSI/N-gI、dR/dt=gI,其中 b 为人均传染率,g 为恢复率。基本再生数 R0=b/g 衡量一名感染者在全易感人群中平均引发的继发感染人数:R0 大于 1 时感染规模初期呈指数增长,R0 小于 1 时疫情自行消退。同一阈值还给出达到群体免疫所需的最低免疫比例 1-1/R0。
The SIR model divides the population into susceptible S, infective I, and removed R classes obeying dS/dt = -bSI/N, dI/dt = bSI/N - gI, dR/dt = gI, where b is the per-capita transmission rate and g the recovery rate. The basic reproduction number R0 = b/g measures how many secondary infections one infective causes on average in a wholly susceptible population: when R0 exceeds 1 the infected count grows exponentially at first, while R0 below 1 lets the outbreak die out on its own. The same threshold yields the minimum immune fraction 1 - 1/R0 required for herd immunity.
难度 进阶 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-11-02
来源 · Source
W. O. Kermack 与 A. G. McKendrick《A Contribution to the Mathematical Theory of Epidemics》,Proceedings of the Royal Society A 115, 1927
W. O. Kermack & A. G. McKendrick, "A Contribution to the Mathematical Theory of Epidemics", Proc. Roy. Soc. London A 115, 1927
独立核验依据 · Independent verification: 核验依据 1 CROSSREF 标识 10.1098/rspa.1927.0118：A contribution to the
#数学建模 #SIR传染病模型
在 SIR 流行病模型中，基本再生数 R0 如何决定疫情的走向？“群体免疫阈值 1-1/R0”又是怎么来的？
In the SIR epidemic model, how does the basic reproduction number R0 determine the course of an outbreak, and where does the herd immunity threshold 1 - 1/R0 come from?
SIR 模型把人群分为易感、感染、移除三类，以微分方程刻画转化。基本再生数 R0 指一名感染者在全易感人群中平均引起的二代病例数：R0≤1 时病例传不满一人，疫情自行消退；R0>1 时病例早期指数增长。令新增感染与移除相抵可知：易感比例跌破 1/R0，即约 1-1/R0 的人口免疫后疫情收缩，这就是群体免疫阈值的由来。
The SIR model sorts a population into susceptible, infectious, and removed classes linked by differential equations. The basic reproduction number R0 is the average number of secondary infections one case produces in a wholly susceptible population: if R0 ≤ 1, each case fails to replace itself and the outbreak dies out, while R0 > 1 gives early exponential growth. Equating new infections with recoveries shows the epidemic shrinks once the susceptible fraction falls below 1/R0 — that is, once a share 1 - 1/R0 of the population is immune, which is precisely the herd immunity threshold.
难度 进阶 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-11-03
来源 · Source
W. O. Kermack 与 A. G. McKendrick“流行病的数学理论贡献”(A Contribution to the Mathematical Theory of Epidemics)，Proceedings of the Royal Society A 第115卷第772期第700–721页，1927，DOI 10.1098/rspa.1927.0111。
W. O. Kermack & A. G. McKendrick, "A Contribution to the Mathematical Theory of Epidemics", Proceedings of the Royal Society A 115(772):700–721, 1927, DOI: 10.1098/rspa.1927.0111.
独立核验依据 · Independent verification: 核验依据 1 CROSSREF 标识 10.1098/rspa.1927.0118：A contribution to the
#流行病建模 #SIR模型 #基本再生数 #群体免疫
为什么疫情前期确诊人数看起来像直线、之后却突然飙升？
Why does an epidemic look like a slow straight line at first, then suddenly explode?
这是指数增长的特征：早期基数小，翻倍所需时间固定，所以绝对增量起初不明显；一旦基数变大，每单位时间新增的人会越来越多，曲线变得陡峭。
This is the signature of exponential growth: the doubling time is constant, so absolute daily additions stay small at first; once the base is large, each doubling adds far more cases and the curve steepens sharply.
难度 入门 · 适合 幼儿(3-6)
CC BY 4.0 · 已核验 · Verified 2026-08-23 · 2 条独立证据
来源 · Source
同济大学数学系 编,《高等数学》第7版,高等教育出版社,2014
Tongji University Dept. of Math, Advanced Mathematics, 7th ed., Higher Education Press, 2014
独立核验依据 · Independent verification: 核验依据 2 ISBN 9787040396638（豆瓣书目条目: 高等数学·上册）
#建模 #指数增长
为什么把利息‘利滚利’比‘单利’长得快得多？
Why does compound interest grow so much faster than simple interest?
单利只在本金上算利息，呈直线增长；复利在‘本金+已生利息’上再计息，呈指数增长。时间一长，二者的差距会被放大到惊人的程度。
Simple interest earns only on principal, growing linearly; compound interest earns on principal plus accumulated interest, growing exponentially. Over time the gap becomes enormous.
难度 入门 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-08-23 · 2 条独立证据
来源 · Source
同济大学数学系 编,《高等数学》第7版,高等教育出版社,2014
Tongji University Dept. of Math, Advanced Mathematics, 7th ed., Higher Education Press, 2014
独立核验依据 · Independent verification: 核验依据 2 ISBN 9787040396638（豆瓣书目条目: 高等数学·上册）
#金融 #指数
疫苗让‘群体免疫’成立的简单数学条件是什么？
What is the simple math condition for 'herd immunity' via vaccines?
当免疫人群比例超过 1−1/R0 时，病毒难以找到易感者持续传播。R0=3 时需约 67% 免疫。这是用基本再生数直接估算的阈值。
When the immune fraction exceeds 1−1/R0, the virus struggles to find susceptibles to sustain spread. For R0=3 about 67% must be immune. A direct threshold from R0.
难度 入门 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-10-04
来源 · Source
《非线性动力学与混沌》Strogatz
Nonlinear Dynamics and Chaos by Steven Strogatz
独立核验依据 · Independent verification: 核验依据 1 ISBN 9780813349107 (2024)
#流行病 #免疫
囚徒困境里，为什么‘都背叛’是稳定的结果？
In the prisoner's dilemma, why is 'both betray' a stable outcome?
对每个人单独看，无论对方怎么选，背叛都是更优的；结果两人都背叛，得到比合作更差的结局。这揭示了个人理性可能导致集体非理性。
For each person alone, betraying is the better move regardless of the other's choice; both then betray and end up worse than if they had cooperated. Individual rationality can yield collective irrationality.
难度 入门 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-11-02
来源 · Source
美国数学会(AMS)《博弈论与经济模型》
American Mathematical Society (AMS), Game Theory and Economic Modeling
独立核验依据 · Independent verification: 核验依据 1 ISBN 9787100047876（豆瓣书目条目: 博弈论与经济模型）
#博弈 #合作
为什么‘半衰期’描述的是指数衰减而不是线性衰减？
Why does 'half-life' describe exponential rather than linear decay?
因为每过固定时间剩余量减半，比例是恒定的，这正是指数衰减：N(t)=N0×(1/2)^(t/T)。线性衰减不会保持固定的‘减半周期’。
Because a fixed time always halves the remainder, a constant ratio, which is exponential decay: N(t)=N0×(1/2)^(t/T). Linear decay would not keep a fixed halving period.
难度 入门 · 适合 大众
CC BY 4.0 · 已核验 · Verified 2026-08-23 · 2 条独立证据
来源 · Source
同济大学数学系 编,《高等数学》第7版,高等教育出版社,2014
Tongji University Dept. of Math, Advanced Mathematics, 7th ed., Higher Education Press, 2014
独立核验依据 · Independent verification: 核验依据 2 ISBN 9787040396638（豆瓣书目条目: 高等数学·上册）
#建模 #衰减