One phone interview question.# DataSciences - 数据科学
m*a
1 楼
本人在PA & DE 交界位置， 各方面都不错，想找一个lover, 有单独的空间，不干扰对
方家庭。 谢谢，非诚勿扰
方家庭。 谢谢，非诚勿扰
D*o
2 楼
Last week in one phone interview, I am asked: in classification, random
forest is very good algorithm, so why do we need other methods?
Now the answer I can imagine is: in some cases, RF may be overqualified. For
example, if the classes are linearly separable, using logistic regression
can give the same accuracy and higher training efficiency.
So anyone has a better answer?
forest is very good algorithm, so why do we need other methods?
Now the answer I can imagine is: in some cases, RF may be overqualified. For
example, if the classes are linearly separable, using logistic regression
can give the same accuracy and higher training efficiency.
So anyone has a better answer?
m*a
3 楼
本人剃毛， 喜欢无毛的
w*d
4 楼
Regulation/ governance 要求，random forest 不行，比如银行
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
w*d
5 楼
Unsupervised clustering 我觉得也可以看成classification ptobelm, RF也不行
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
w*d
6 楼
AlphaGo 估计也不是RF
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
C*t
7 楼
Interpretability?
N*N
8 楼
支持这个，另外就是rf也不一定是最好的，boosting在有的数据上会更好
也可以提一下这个：
https://en.wikipedia.org/wiki/No_free_lunch_theorem
【在 C**********t 的大作中提到】
: Interpretability?
也可以提一下这个：
https://en.wikipedia.org/wiki/No_free_lunch_theorem
【在 C**********t 的大作中提到】
: Interpretability?
a*e
9 楼
Sometimes you need to explain the model
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
t*k
10 楼
people will prefer a model with 90% accuracy but tell you how and why than a
RF model with 95% accuracy but only tell you how
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
RF model with 95% accuracy but only tell you how
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
s*h
11 楼
1.速度啊速度。
2.解释啊解释
3.软件支持
4.不够好啊不够好。NN beat RF的 情况很多的。
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
2.解释啊解释
3.软件支持
4.不够好啊不够好。NN beat RF的 情况很多的。
For
【在 D****o 的大作中提到】
: Last week in one phone interview, I am asked: in classification, random
: forest is very good algorithm, so why do we need other methods?
: Now the answer I can imagine is: in some cases, RF may be overqualified. For
: example, if the classes are linearly separable, using logistic regression
: can give the same accuracy and higher training efficiency.
: So anyone has a better answer?
s*n
12 楼
这是为什么？监管者为什么要管银行用什么算法？
是指这模型不能给出明确insights么，可能在未来出问题么。
【在 w******d 的大作中提到】
: Regulation/ governance 要求，random forest 不行，比如银行
:
: For
是指这模型不能给出明确insights么，可能在未来出问题么。
【在 w******d 的大作中提到】
: Regulation/ governance 要求，random forest 不行，比如银行
:
: For
l*n
13 楼
fed，occ的report都是有规定的，你只能在他们规定的框架内干事情，这就是
regulation， no why question.
【在 s***n 的大作中提到】
: 这是为什么？监管者为什么要管银行用什么算法？
: 是指这模型不能给出明确insights么，可能在未来出问题么。
regulation， no why question.
【在 s***n 的大作中提到】
: 这是为什么？监管者为什么要管银行用什么算法？
: 是指这模型不能给出明确insights么，可能在未来出问题么。
相关阅读
请问想找data scientist 工作应该怎么开始准备求大侠指点大数据课题预算请问学习 d3.js 有什么好的参考资料吗？OPT extension问题免费视频， tableau新开讲Data Scientist Intership at a big network company, south bay area, california《行业知识》系列讲座 (转载)也许这个问题有点模糊。如何设计建立一个Hadoop Cluster？都应该从哪些角度考虑？做credit risk scorecard的朋友们， 请进来， 有问题求教 (转载)uber analytics test转行数据挖掘和机器学习拿到两个offer,求建议怎么处理categorical variable有很多个level的寻找统计相关工作 data analyst/scientist请问决策树连续值的分界点怎么选问个matching algorithm免费讲座 BI/Data Analyst 就业市场分析 (转载)google data scientist/quantitative analyst有什么模型能把linear regression model 和 time series model (转载)Voice recognition请教