GroupLens: A group good at Recommendation

We can find articles and data sets there, very useful site: http://www.grouplens.org/

21st Canadian AI (2008) CFP

Full paper submission due January 15th, 2008
Notification of acceptance February 26th, 2008
Final paper due March 15th, 2008

Begin to learn Lucene

Web and text mining is what I am interested but haven't touched much. Lucene is a good starting point since 1) easy basic idea to understand; 2) open-source code to study things happening behind; 3) search technology is hot and basic for more advanced text mining.

Excellent source about data mining jobs

Click here!

SUMM: Something to compete with Conjoint Analysis

SUMM vs. Conjoint Analysis

• SUMM can handle three or four times as many attributes as Conjoint because of its unique measurement scale.
• SUMM incorporates each respondent's subjective beliefs about vendors in each choice simulation, while Conjoint requires each product to be defined objectively.
• SUMM flows from a straightforward theory of how people make choices, so it is easier to understand how the numbers are generated.

You can find more about SUMM on Eric Marder Associates.

Powerful Conjoint Analysis

Conjoint analysis provides useful results for product development, pricing research, competitive positioning, and market segmentation. It also measures brand equity.

1st Educational Data Mining (EDM) (2008)

First International Conference on Educational Data Mining

Data Mining and Statistics in Service of Education

Call for papers (preliminary)

http://www.EducationalDataMining.org

June 20-21, 2008

Co-located with International Conference on Intelligent Tutoring Systems (ITS 2008)

The First International Conference on Educational Data Mining brings together researchers from computer science, education, psychology, psychometrics, and statistics to analyze large data sets to answer educational research questions. The increase in instrumented educational software, as well as state databases of student test scores, has created large repositories of data reflecting how students learn. The EDM conference focuses on computational approaches for using those data to address important educational questions. The broad collection of research disciplines ensures cross fertilization of ideas, with the central questions of educational research serving as a unifying focus. This Conference emerges from preceding EDM workshops at the AAAI, AIED, ICALT, ITS, and UM conferences.

Topics of Interest

We welcome papers describing original work. Areas of interest include but are not limited to:

· Improving educational software. Many large educational data sets are generated by computer software. Can we use our discoveries to improve the software’s effectiveness?

· Domain representation. How do learners represent the domain? Does this representation shift as a result of instruction? Do different subpopulations represent the domain differently?

· Evaluating teaching interventions. Student learning data provides a powerful mechanism for determining which teaching actions are successful. How can we best use such data?

· Emotion, affect, and choice. The student’s level of interest and willingness to be a partner in the educational process is critical. Can we detect when students are bored and uninterested? What other affective states or student choices should we track?

· Integrating data mining and pedagogical theory. Data mining typically involves searching a large space of models. Can we use existing educational and psychological knowledge to better focus our search?

· Improving teacher support. What types of assessment information would help teachers? What types of instructional suggestions are both feasible to generate and would be welcomed by teachers?

· Replication studies. We are especially interested in papers that apply a previously used technique to a new domain, or that reanalyze an existing data set with a new technique.

Important Dates (tentative)

· Paper submissions: March 31, 2008

· Acceptance notification: April 30, 2008

· Camera ready paper: May 16, 2008

· Conference: June 20-21, 2008

Submission types:

· Full papers: Maximum of 10 pages. Should describe substantial, unpublished work.

· Young researcher: Maximum of 8 pages. Designed for graduate students and undergraduates.

Promote one site on the research on Conjoint Analysis

http://conjointonline.com/

Welcome to post more good resource here.

Thanks!

Spatial Database and Data Mining

Spatial data mining seems an emerging application, attracting more and more researchers. I find this group in UMN has interesting work, http://www.spatial.cs.umn.edu/, and you find some overall information from their survey article, http://www.spatial.cs.umn.edu/paper_ps/dmchap.pdf

Normalized Google Distance (NGD)

It is interesting to measure the two terms via information from Google. see paper

Paper accepted by Australian AI 2007

My paper, "Local Learning Algorithm for Markov Blanket Discovery: Correct, Data Efficient, Scalable and Fast", that co-authored with my professor Michel Desmarais is justed accepted as full paper by Australian National Conference on AI 2007. Keep working for more!!

What is Hierarchical Bayes

Hierarchical Bayes models are actually the combination of two things: 1) a model written in hierarchical form that is 2) estimated using Bayesian methods. Its high-demanding in computing resource prevents its practice. Fortunately, this status is turning better upon the introduction of MCMC sampling technique.

Standards to evaluate Conjoint Analysis

1. Low respondent burden (including width and length of the survey);
2. High accuracy in prediction to the survey initiator.

CFP: FLAIRS-21(2008)

Deadline for the paper submission date is 19th November, 2007. The conference will be hold from 15th to 17th May, 2008. (official link)

Two new terms met recently

"Conjoint Analysis" and "Hierarchical Bayes" are the two I never heard of till very recently. Both of them appear in marketing research community. Looks interesting topics, so I will pay attention on them before I can figure out more with you.

D-separation Tutorial

You can find the history and D-separation from http://www.andrew.cmu.edu/user/scheines/tutor/d-sep.html#d-sepapplet2, where, a Java applet is provided to assist you learn this concept.

Computational Intelligence

Computational intelligence (CI) is a successor of AI. As an alternative to GOFAI it rather relies on heuristic algorithms such as in Fuzzy systems, Neural Networks and Evolutionary computation. In addition, computational intelligence also embraces techniques that use Swarm intelligence, Fractals and Chaos Theory, Artificial immune systems, Wavelets, etc.

One research center about CI: http://www.cs.vu.nl/ci/

Markov Equivalent Class Graph

A Markov equivalence class can be represented by a graph with
1. Same link and
2. Same uncoupled head-to-head meetings
Any assignment of directions to the undirected edges in this graph, that does not create a new uncoupled head-to-head meeting or a directed cycle, yields a member of the equivalence class.

(Find detail on http://mmc36.informatik.uni-augsburg.de/mediawiki/data/3/3a/SS07_MM2_Lec06-Bayesian_Networks_chapter02_Part2.pdf )

Good tutorial of Support Vector Machine

http://www.umiacs.umd.edu/users/joseph/support-vector-machines4.pdf

CFP: AIKED'08

Submission deadline: November 15, 2007
Acceptance notification: December 31, 2007
Conference date: February 20-22, 2008
Location: Cambridge University

AIKED'08 = The 7th WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Data bases