19.txt

This complete matlab for neural network

发信人: NAOMIELIE (雁来红), 信区: DataMining
标  题: Re: 大家谈谈数据?掘最新的国内外研究现状!
发信站: 南京大学小百合站 (Fri Mar 28 23:41:25 2003)

I do not think data mining will be useless in real-life applications.
And, in fact, those professors do have their own things to do,
at least those famouse ones.

From applications aspect, they DO need data mining.
(this does not include those enterprises in China, 
which are outliers.)
You can learn this from those banks, telecommunication companies,
consulting companies, and other large companies.

From research repective, 
firstly, you should have your own background.
Data mining problems are raised by users,
and you just use your own way to resolve these problems.

You can see that ai people are continuing building more general models,
statistics are trying to remove the assumptions so that their methods can
be applied on real-life complex data. 
And database people are trying to integrating data mining operations into
databases, so that they are more efficient and ease-of-use.

A professor in hkust told me last week that he never worried about problems
as 'what to do next', since, firstly,
there are quite a lot of companies are searching for data mining people,
and, he has strong background in planning, which is a research domain of ai.

Furthermore, he thinks he knows the 'direction' of applications,
which is distribution, autonomos, and mobil.

For research students, I think the problem is that you must have some strong 

background, in either statistics or cs (I don't think people from 
application domains can do very good data mining research, although
they may make very successful data mining application cases/stories.).

So, if you come to data mining domain with empty hands,
you may get nothing.
And, if you have something, you needn't worry about anything.


【 在 intel 的大作中提到: 】
: 国内的数据?掘很热,但国外的情况是每年开kdd,大家都在说下面该做什?了
: 有人说data mining 可能会想80年代的expert system一?,学?界多说好,有前途,
: 但是?际上没人用.
: 以上?老板所言
: 个人的?点:找个?际的东西,如CRM,Robot,具体做一下,可能会?现很多问题的是
: 值得研究的
: 【 在 kailey (kailey) 的大作中提到: 】
: : 
: : 看了一些?于数据?掘国内外研究现状的资料,大部分都是一个版本,
: : 没有找到?于最近几年数据?掘的国内外研究现状,请大家谈谈!
: : 谢谢!

--

※ 来源:．南京大学小百合站 http://bbs.nju.edu.cn [FROM: 210.172.103.218]