Subscribe to DSC Newsletter

I am trying to decide which program to apply for this summer and was wondering if anyone has any knowledge or experience with the content/scope of any of the programs I have on my list. I have 17 years statistical consulting experience, a Masters in Statistics, and SAS/STATISTICA/R skills. Any advice on which program might be best for my background? I ruled out the OSU program because I wanted more of a computer science vs marketing focus. I would like to get some hands on experience with Hadoop, web mining, text mining, Python or Perl, but the descriptions in most of these programs don't indicate that as part of their content. 

- Stanford
- UC San Diego Extension
- Central Connecticut State University
- University of Louisville
- New Jersey Science and Technology Institute

Thank you in advance for any information you can provide on the subject,
Sandra Donlon

Views: 4594

Replies to This Discussion

Hi Sandra - I've done a similar search and also been a little disappointed in what's available within the scope of "data mining" certificate or masters programs. University of Louisville's program looks a lot like OSU's to me with the focus being mostly on operating SAS Enterprise Miner rather than deeper learning. Stanford's certificate program also looks pretty fluffy to me.

You may have more luck finding what you're looking for by concentrating on machine learning rather than data mining , particularly if you're interested in parallel computation and computer science. Here are a few of the programs I'm aware of in that vein:

Machine Learning @ Carnegie Mellon University:

*Full Disclosure: I went to CMU as an undergrad so I'm biased. They only offer a PhD or secondary masters for now, but the faculty are top notch. Cosma Shalizi and Tom Mitchell are fantastic and the only reason I know anything about data mining.

Computer Science & Engineering @ University of Washington

(They offer a master's degree for professional's with courses on parallel computing, AI, data mining, etc.)

Machine Learning @ University College London

I'm sure there a number of other ones out there, these are just ones I'm aware of. If you find any programs in your search that are a nice balance of theory & application of data mining and machine learning - let me know! I'd be interested.

Good luck!

I've taken 9 out of the 10 courses for the MS in Data Mining at
CCSU. It is the only online MS DM program, as far as I know.
The MS requires 10 courses and a masters thesis. The certificate
program requires the three DM methods courses STAT 521,522,523
and two out of three of these other courses: Text Mining (STAT 527), Web Mining (STAT 525), Genomics (STAT 526.
The methods courses are taught out of Dr. Larose's textbooks
(Discovering Knowledge in Data ; Data Mining Methods and Models), and we CCSU
students buy SPSS Clementine (now PASW Modeler after IBM acquisition) for $200
for the GradPack. The software development environment is very friendly in Clementine.

Text Mining is Perl-based and uses Dr. Bilisoly's text. See

The Web Mining course uses a text by Drs. Larose and Markov (also at the above website).
The course focuses on web log analysis.

The other four courses I had to take were:
Probability and Statistical Inference (STAT 416) (you've surely mastered this stuff);
Multivariate Statistical Analysis (STAT 570) (text by Johnson), which assumes you know Linear Algebra -
this course is necessary to really grasp what multivariate analysis really is; Uses SPSS Statistics,
although SAS students used their packages with no problem;
Regression Analysis and Time Series (STAT 567);
Machine Learning (CS 580), which is the theoretical heart of DM. The text is Tom Mitchell's of CMU, along with Prof. Markov's material.

I feel that the CCSU program has served me very well. Dr. Larose makes us do our projects as if to actual clients. That focus enabled me to do a Machine Learning project that led to a great job.

with best wishes,
Ben Dickman
I need help or literature that will help in constructing research proposal in the application of data mining in drug designs/clincal trials/bioinformatics. I will like to incorporate the efficiency of half-normal plots in my research and this is not for undergraduate.
I agree with you Sandra those programs do not give enough info/details. The University of Louisville and OSU are similar in that they are training grounds for SAS. This of course is not necessarily a bad thing depending on your goals.

I myself was looking for a program with mathemtical/computer science content especially coming from a Masters in Engineering. However the programs mentioned by Jason while satisfying that requirement are too expensive.

Machine Learning @ Carnegie Mellon University
Machine Learning @ University College London

The Certificate at CCSU looks good but right now is outside of my budget and time schedule. I do like the three application courses in Web Mining, Text Mining and Genomics

I am also do not want to spend entire year etc. learning data mining. My postgraduate degree already involved a Masters with Neural Networks and Genetic Algorithms etc.

I prefer something shorter, a semester at the maximum but also not so expensive that I will have to take a loan to complete.

Therefore for myself I have come to the conclusion that SAS training courses which last two to three days at the most in Data Mining Techniques, Predictive Modeling etc., Web Mining and Customer Intelligence will be the most useful training. This will give me domain or real world applied knowledge.

The mathematical and applications components I am looking for will come from mainly two books Handbook of Statistical Analysis and Data Mining Applications by Robert Nisbet, Gary Miner (less maths) and The Elements of Statistical Learning: Data Mining, Inference, and Predictionby Trevor Hastie, Robert Tibshirani, Jerome Friedman (more maths).

I can teach myself anything that is necessary from these two books. After all it is only maths.

This is the most affordable option for me.



On Data Science Central

© 2020 is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service