The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was no consensus. T. Canens (talk) 03:02, 13 June 2010 (UTC)[reply]

Environment for DeveLoping KDD-Applications Supported by Index-Structures

[edit]

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

Fails to meet WP:Notability. Contested PROD, tagged for ((notability)) since 2 May with no addition of references to indicate significant coverage in reliable sources that are independent of the subject. Searching Google Scholar, Books and News finds none either. Current references are all written by the developers. Qwfp (talk) 17:23, 29 May 2010 (UTC)[reply]

Note: This debate has been included in the list of Software-related deletion discussions. —Qwfp (talk) 17:25, 29 May 2010 (UTC)[reply]

The current references are all in established, highly regarded, peer-reviewed scientific publications, not self-published or pay-for-publish. Quoting WP:RELIABLE:

Material such as an article or research paper that has been vetted by the scholarly community is regarded as reliable. If the material has been published in reputable peer-reviewed sources or by well-regarded academic presses, generally it has been at least preliminarily vetted by one or more other scholars.

[1] has SSDBM and DASFAA in the Rank 2 group for Databases. SSTD/SSD is a specialized conference for time series and spatial databases, only every other year, but also peer-reviewed and highly regarded in the research community.

The software has been used for visualization in R-Tree, Local Outlier Factor, de:OPTICS (OPTICS algorithm) and I intend to use it for visualization of DBSCAN.

Keep, but I am the primary author of the article, and a contributor to the software --Chire (talk) 18:10, 29 May 2010 (UTC)[reply]
Delete - sources given do not establish notability of the software, merely its existence and features.--70.80.234.196 (talk) 19:19, 29 May 2010 (UTC)[reply]
Comment - First of all the article should be named ELKI as an acronym to the very long name currently used. Sources are from established conferences and they could establish notability but NOT neutral point of view (they are publications from the team that created the software). Since the article's author has a vested interest in the software there are additional problems with neutrality. I would suggest the authors to try to find additional resources and references from other authors to help establish the above point and get other wikipedia editors to improve the neutrality of the article. And please change the article's name... Pxtreme75 (talk) 10:04, 30 May 2010 (UTC)[reply]
Comment - I tried to keep the article as neutral and factual as possible. If you find anything non-neutral, you are welcome to help improving it. As for the name: it may be ugly, but at least its collision-free. And it is the full official name, capitalization included. I did not want to introduce the article as ELKI because it might at some point be needed to have a disambuigation page then, and "grabbing" short names would be "non-neutral" in my opinion. Following WEKA to Weka (machine learning), it might however be okay to use ELKI (data mining)? --Chire (talk) 12:19, 30 May 2010 (UTC)[reply]

Comment. Neutrality is achieved from proper use of different types of sources. My problem is not with the tone of the article but the absence of multiple points of view (check Verifiability). Are there articles from other authors that use ELKI? Any non-biased reviews on established web or printed mediums? On the other topic ELKI (data mining) seems a reasonable choice. Pxtreme75 (talk) 18:28, 30 May 2010 (UTC)[reply]

Comment: it is independently acknowledged here, but at that time (0.1) deemed not as well-suited as WEKA for their requirements (no visualization yet at that time). There is this master thesis indepently done in Greece using ELKI source code (but it's just a masters thesis, so forget it as reference). There is an upcoming publication to be presented next week at DaMoN2010 that used ELKI for experiments with spatial indexing on solid-state disks [2] using the R*-tree implementation of ELKI, however that one is not entiely independent (colleagues of the ELKI authors). There is [3], a joint work with the University of Alberta, Edmonton, Candada, that only includes one of the ELKI contributors, but thus it's also not independent. There is this citation analysis on databases that asserts the database research group to be the 10 most cited worldwide (but there is no page on the database group that the article could be merged into). There is an upcoming workshop at the ACM KDD2010 conference in the MultiClust track that will likely involve ELKI (given that two of the contributors are involved). But seriously, I don't have the impression any of this will satisfy your requests for independent evaluation (although all cited references are peer-reviewed, usually by 3-4 domain experts) despite WP:OBSCURE. I doubt you'll ever see this software reviewed in a traditional print magazin or web software review site, given that it's research software and not point-and-shoot data mining software (if such things even can exist). Deletionism on Wikipedia is driving me nuts. --Chire (talk) 07:54, 31 May 2010 (UTC)[reply]
Additional reference: ACM SIGKDD 2009 Dissertation Award Runner-Up for correlation clustering. The two-page summary can be found on kdd.org. Quoting: "The implementations of all algorithms are available in the framework ELKI [8; 1]" (the recipient of the award is one of the main authors of ELKI. The two references are the ELKI 0.1 and 0.2 releases). So when the #1 data mining conference gave a dissertation award to it, does this finally establish enough WP:NOTABILITY for an WP:OBSCURE software page? (and can we then get back to working on content again?) --Chire (talk) 18:52, 31 May 2010 (UTC)[reply]
Comment. First of all I don't believe in either deletionism or inclusionism. The truth usually lies somewhere in between. I do believe though that all these rules can turn against the truth. So, I prefer to judge things case by case. The WP:OBSCURE nature of the software consists no problem for its inclusion. Also, personally I am not sure if it merits an article or not. That is why i have not voted - I just offered some comments and suggestions. Pxtreme75 (talk) 21:44, 1 June 2010 (UTC)[reply]

Relisted to generate a more thorough discussion so consensus may be reached.
Please add new comments below this notice. Thanks, JForget 00:49, 5 June 2010 (UTC)[reply]

Weak keep- The article does have scholarly references. I also believe the article needs to be renamed, the current title is unwieldy.Smallman12q (talk) 02:16, 5 June 2010 (UTC)[reply]
The references are sound; I have much less concerns about COI than e.g. with RapidMiner and KNIME. These four (with Weka) all seem pretty similar, with only Weka really standing out because of book coverage in machine learning (it also is used by KNIME and RapidMiner, which seem to be frontends, while ELKI seems to aim for becoming the "Weka in data mining" instead of machine learning?). That probably is the reason why the Weka article is a lot better. Oh, and I'm all in for renaming to ELKI (data mining), the current page name is awful. --87.174.80.103 (talk) 11:42, 6 June 2010 (UTC)[reply]
P.S. reminds me a bit of Wikipedia:Articles_for_deletion/Weka_(machine_learning). We should maybe see it as research, not software. --87.174.80.103 (talk) 11:50, 6 June 2010 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.