Search Results Clustering experiment with Ambient and ODP239 using original Lingo algorithm

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Search Results Clustering experiment with Ambient and ODP239 using original Lingo algorithm

seyfullahd
Hi,

Since ODP239 and AMBIENT datasets were firstly introduced later than when your Lingo and Carrot2 related papers published, naturally there is no experiment results with these datasets using original Lingo (with SVD factorization) algorithm. And, I have no chance to set up and experiment using these datasets and original Lingo algorithm. I wonder if you can set up an experiment for this and share results. (At first I thought I can acquire original Lingo algorithm by modifying current source code like changing parameter values and threshold values. And then I thought it would never be the original Lingo algorithm :) Therefore, I suppose there is no way to acquire original algorithm and reexperiment.)

Secondly, as far as I can understand from the papers you used some randomly chosen major and outlier categories from ODP, for experiments. I wonder, if I can get detailed results like

* For [1] and [2] (same experiment), clusters with their labels and snippets assigned to each clusters, and also ground truth original partitions with their categories and snippets (since, now they could be changed) for original Lingo algorithm.
* For [3], same results for Lingo (NMF-ED) algorithm

Thanks,
Seyfullah

REFERENCES:
[1] - Stanisław Osiński, Dawid Weiss: Conceptual Clustering Using Lingo Algorithm: Evaluation on Open Directory Project Data. Advances in Soft Computing, Intelligent Information Processing and Web Mining, Proceedings of the International IIS: IIPWM´04 Conference, Zakopane, Poland, 2004, pp. 369—378
[2]- Stanisław Osiński, Dawid Weiss: A Concept-Driven Algorithm for Clustering Search Results. IEEE Intelligent Systems, May/June, 3 (vol. 20), 2005, pp. 48—54
[3] - Stanisław Osiński: Improving Quality of Search Results Clustering with Approximate Matrix Factorisations. Springer Lecture Notes in Computer Science, vol. 3936, pp. 167—178, Proceedings of the 28th European Conference on IR Research (ECIR 2006), London, UK, April 10-12, 2006
Reply | Threaded
Open this post in threaded view
|

Re: Search Results Clustering experiment with Ambient and ODP239 using original Lingo algorithm

Stanislaw Osinski-3
Hi,

Since ODP239 and AMBIENT datasets were firstly introduced later than when
your Lingo and Carrot2 related papers published, naturally there is no
experiment results with these datasets using original Lingo (with SVD
factorization) algorithm. And, I have no chance to set up and experiment
using these datasets and original Lingo algorithm. I wonder if you can set
up an experiment for this and share results. (At first I thought I can
acquire original Lingo algorithm by modifying current source code like
changing parameter values and threshold values. And then I thought it would
never be the original Lingo algorithm :) Therefore, I suppose there is no
way to acquire original algorithm and reexperiment.)

I'll search my old archives, but this is 10 years, several laptops and versioning system changes later, so the original code may not have survived this :-)

 
Secondly, as far as I can understand from the papers you used some randomly
chosen major and outlier categories from ODP, for experiments. I wonder, if
I can get detailed results like

* For [1] and [2] (same experiment), clusters with their labels and snippets
assigned to each clusters, and also ground truth original partitions with
their categories and snippets (since, now they could be changed) for
original Lingo algorithm.
* For [3], same results for Lingo (NMF-ED) algorithm

Same issue here. Later today I'll try to search my archives for the code and the results.

S.

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_______________________________________________
Carrot2-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/carrot2-developers