Talk:Educational data mining

Merge from Cluster analysis

Cluster analysis is overfull and has lots of text on the use of cluster analysis in educational research, that should better go here. --87.174.68.114 (talk) 20:38, 26 September 2011 (UTC)

Done so, makes a lot of sense. --Chire (talk) 08:13, 3 October 2011 (UTC)

Inappropriate text removed

An inappropriate long section on cluster analysis, the vast majority of the text irrelevant to the topic of educational data mining, was removed. Cluster analysis is one area of EDM, among many; having over half the article dedicated to one article describing a single study, plus great detail on that one area, unbalances the article. Ryan22222 (talk) 03:11, 4 October 2011 (UTC)

Off-topic: please learn how to use the discussion pages properly. There are important markers there that should remain at the top, you are expected to add your comments (for example using the "new section" button) to the end. Thank you. --Chire (talk) 20:49, 5 October 2011 (UTC)

Secondly, the content here was moved over - which is well documented in the changelogs and on this very talk page - from another article. It was there misplaced also, which is why I moved it here. I'm all with you that it it too long and imbalanced. However removing it in its entirety is not acceptable in my opinion. Go ahead, shorten it to the relevant parts. Or even better: elaborate on the stuff that you consider is underrepresented now, because the article is far from perfect in the other parts, too (heck, many of the links don't even work, yet to speak of the non-encyclopedic bullet point style). But just completely dropping the section is an offensive action against some other editors work, which I really consider inacceptable. Fix it, don't delete it! Thank you. --Chire (talk) 20:49, 5 October 2011 (UTC)

Respectfully, around 90% of this section added in from elsewhere is off-topic or out of date. In a fast-moving new field, citations from the 1970s aren't really appropriate, and many of the newer citations are not known within this research community. You've said in a private message that as a researcher who publishes in this field, and who is cited in the current page, I have a conflict of interest and should not edit this page. I was not aware this was wikipedia's policy, but I will ask colleagues who publish in the field, at other universities, and who are not currently cited, to edit. I agree that the previous version of the page was not top-quality, but adding large amounts of low-quality/irrelevant content is unlikely to make for a better page. Ryan22222 (talk) 21:59, 5 October 2011 (UTC)

I did not say you shouldn't edit at all. In fact, in my (very personal) opinion Wikipedia needs the knowledge of researchers. And you might be much more qualified than the author of that section. But you shouldn't just blank entire sections (please do read this page!) -- and having a friend of yours blank the section as it happened just now is not at all better. This is called a Wikipedia:Meatpuppet. Please leave the section in (see Wikipedia:PRESERVE), tag it with appropriate templates (as the one I put in) until you find the time to replace it with better content. If 90% of the section are off-topic and out of date, remove just these 90%. This is what makes a better page; not removing any content from Wikipedia you do not like. This will make an empty wikipedia only. Thank you. --Chire (talk) 23:22, 5 October 2011 (UTC)

Chire, with all due respect, your text makes a false accusation. You say that I "had a friend of mine blank the section" and referred to this as a Wikipedia:Meatpuppet. Your previous message, sent personally to me (rather than posted on this page), appeared to tell me I was not allowed to edit. I may have misunderstood you. But for this reason, I emailed several colleagues (across universities, chosen for their senior position in the field rather than close relationships to me) to ask them to edit. One of them appears to have come to the same conclusion that I originally did, that the text should be removed. I can see who it was from the login. He's certainly not my "meat puppet". He's at least as senior a figure in the field as I am, we've never written an article together, and his 2007 review of this area is significantly more well-cited than the review of the area I've published. Other colleagues chose instead to comment on the page. As of the time I'm writing this, this page now has at least four widely-cited researchers in this field, representing three different universities, who believe this text is inappropriate for this page. Ryan22222 (talk) 02:54, 6 October 2011 (UTC)

Sorry, this was not meant to offend you or your colleagues. If you investigate edit patterns in Wikipedia, you will see that on controversial edits, these kind of patterns happen (which is why there is a Wikipedia document on the issue), and should thus be avoided in "legitimate" consensus finding. However, the consensus should ideally consist of more than "undo the work of User:Jucypsycho", but for example in: "yeah, we should expand on the areas, with one paragraph each. We're shortening his section to this paragraph, and add other paragraphs for the other areas later on." --Chire (talk) 07:22, 6 October 2011 (UTC)

I think part of the issue is whether this content belongs on this page or somewhere else. This page has largely been the province of the community who publishes at the Educational Data Mining conference. Does that community find this "Schools Like Mine" cited throughout the pasted section to be representative of "Educational data mining"? If not, would Learning analytics or Academic analytics (latter linked from the former) be more appropriate? Turadg 01:02, 6 October 2011 (UTC) — Preceding unsigned comment added by Turadg (talk • contribs)

Yes, merging the content into a page that is more appropriate or a page of its own is an excellent option. I'd oppose a page of its own (I don't see enough relevancy for that), I'd personally even delete it then; but such a change should be discussed first. This discussion so far wasn't much about the content changes, but much more about the way these changes are performed. Dropping this section some other editor had put in a lot of effort is quite disrespectful, which is why I object to it.

If the article is meant to be about that particular research community, then indeed the section doesn't fit in. But then the article should be structured and written that way that it explains the communitys structure. And maybe even have working links to the community conferences... ;-) --Chire (talk) 07:22, 6 October 2011 (UTC)

I don't think that the issue is whether the page represent a certain community or not but about clarity of the page and appropriate links to other pages (as the page does already --- I imagine broken links are fixed now as I tried them and all work). About the Clustering content that started this discussion topic, you can imagine that the same thing would happen with content from other techniques, making the page pretty unreadable; I don't think it is about blanking sections on purpose, or inappropriate referencing or not. I suggest that to resolve this issue instead of dumbing text from somewhere else to make it less clutter here (as discussed below) new pages are created with subject e.g. Cluster Analysis in Education (or ERM or Learning Analytics for that matter ;-)). I would find this very interesting and even contribute to such page(s). Hope this helps. --95.146.161.142 (talk) 09:06, 7 October 2011 (UTC)

The Cluster Analysis stuff is way too much and should be deleted wholesale, or reduced to a sentence as one of dozens of EDM technigues

Someone has tagged the article with "major contributor to this article appears to have a close connection with its subject". Ryan Baker happens to be the duly elected president of the Educational Data Mining Society, so "scare" notice seems inappropriate. He is also the author of the Handbook of EDM. See http://www.kdnuggets.com/2010/11/handbook-educational-data-mining.html I will say, I know the gentleman, and at times have co-authored papers with Baker (In fact if you don't know who the president of the educational data mining society is, that maybe should that give someone pause before editing the article. )

Don't we want the experts in the field helping to make the best articles?

I am no expert on how to write wikipedia articles but I am an expert in educational data mining. No only have I had paper accepted at every Educational Data Mining conference since it began, my student won Best Paper in past years. I had 4 papers at the conference in 2011. I also write papers that use clustering. For instance

Trivedi, S., Pardos, Z., Sarkozy, G. & Heffernan, N. (2011) Spectral Clustering in Educational Data Mining. In Pechenizkiy, M., Calders, T., Conati, C., Ventura, S., Romero , C., and Stamper, J. (Eds.) Proceedings of the 4th International Conference on Educational Data Mining. Pages 129-138.

and Trivedi, S., Pardos, Z. & Heffernan, N. (2011) Clustering Students to Generate an Ensemble to Improve Standard Test Score Predictions In Biswas et al (Eds) Proceedings of the Artificial Intelligence in Education Conference 2011. Springer. LNAI 6738, Pages. 328–336. http://users.wpi.edu/~zpardos/papers/AIED2011_clustering_ensemble.pdf

So I think I am in a excellent position to judge the appropriateness of the large section on clustering. But I must say that that huge section on clustering does not belong. I read on the discussion board that an editor suggested "fixing" it rather than deleting so I am an not sure what to say. I thinking the whole thing should be deleted. I decided to now do that YET as a discussant suggested that is the wrong way of going about it. But the idea that adding details to other section is probably not a good way of dealing with this. To me that section seems so out of place.

I know we should "assume good intent" on behalf of others but .... I don't want to make this any more personal than it has already become but if we are going to attack the president of the society as being bias, what is the background of the person suggesting the clustering work deserves to be there.

Is it someone that is an expert clustering and want to push that work into all the different venues it could possibly relate to? This is what is seems like to me.

And all of this started cause "Cluster analysis is overfull and has lots of text on the use of cluster analysis in educational research, that should better go here. --87.174.68.114 (talk) 20:38, 26 September 2011 (UTC)"? So someone thought text did not belong there so they dumped it here? I don't see why this is anyway to do deal with this. If they don't want the text why should EDM have to accept it? Dumping it into a different article sounds good to me but I assume the right thing is to delete it as I don't think its very good. Why should they have to go through the same stress as this community has to go through?

Neiltheffernaniii (talk) 01:12, 6 October 2011 (UTC)

COI, inappropriate blanking, and referencing

This article has been the subject of an enquiry at WP:EAR. Nobody owns Wikipedia articles whoever creates them or adds content. It may be possible for people with a conflict of interest to edit th encyclopedia provided they have read and fully underestood the policy which is mainly directed at maintaining neutrality and preventing editors from promoting themselves, their organisations, or anyone or anything they are closely related to. It seems here that employees may have technical knowledge that may (I'm not an expert on the subject) beneficially expand the article. However, an encyclopedia must deal with facts, and information that is up to date, and the content must be referenced. Please continue your discussion here, but keep it on a civil note. Blanking sections of the article, or blanking sections of relevant discussion on your user talk pages will only exacerbate the issue. Thank you. --Kudpung กุดผึ้ง (talk) 04:20, 6 October 2011 (UTC)

Page protected

The disruption has not stopped so I've sem-protected the page so that it can only be edited by registered, established users. If the problems persist, I'll protect it for admin editing only. --Kudpung กุดผึ้ง (talk) 13:14, 6 October 2011 (UTC)

Suggestions

This may just be me, but I'm confused by "Domain Definition" versus just a "Definition" section? What is the difference between the two? This article may benefit from a reference to Big data as from reading this I understand that this article is related directly to, but this just focuses on education. Perhaps some additions to the applications and examples like some explanations or links would be beneficial to how the article is understood. SII&CT Alex (talk) 20:43, 27 February 2014 (UTC)

Thanks for your suggestions, SII&CT Alex! It seems to me that "domain definition" refers to a definition that is controlled by those that contribute to that particular area of knowledge. I agree that this is confusing from just "definition. Since I am expanding on this existing EDM article, I will leave this comment for the contributor of that section to address. Contextualizing this topic through the applications and examples, and references to big data, are great suggestions which I hope to incorporate. --Mabelho (talk) 14:35, 30 March 2014 (UTC)