{ "id": 11409, "title": "Links on identification sheet: do they commute?", "url": "/competitions/datasciencebowl/discussion/11409", "author_name": "", "post_date": "2015-01-01T20:24:29.553Z", "votes": null, "comment_count": 4, "views": 1132, "content": "

Hi,

\n

Just wondering if anyone knows whether the dotted lines on the plankton identification pdf commute or not.  That is:  there are links between four of the echinoderm classes, for example.  Are they all alike, or would there be more lines between them if they were?

\n

Thanks!

\n

-J.Q.

", "messages": [ { "id": "62187", "postDate": "01/01/2015 20:24:29", "content": "

Hi,

\n

Just wondering if anyone knows whether the dotted lines on the plankton identification pdf commute or not.  That is:  there are links between four of the echinoderm classes, for example.  Are they all alike, or would there be more lines between them if they were?

\n

Thanks!

\n

-J.Q.

", "rawMarkdown": "", "votes": null }, { "id": "62372", "postDate": "01/06/2015 17:50:23", "content": "

Hi J.Q.,

\n

The dotted lines mean that there is some feature-related relationship between the groups. They're separate groups (not all alike), but similar to each other in some way. They are not derivative of each other in either taxonomy or shape, or there would be solid lines between them. 

\n

Hope this helps,

\n

Jessica

", "rawMarkdown": "", "votes": null }, { "id": "62374", "postDate": "01/06/2015 18:20:41", "content": "

Hi Jessica,

\n

Thanks.  So, just to be clear, we have, for example, 

\n

echinoderm larva pluteus early

\n

<----->

\n

echinoderm larva pluteus urchin

\n

<------>

\n

echinoderm larva pluteus star

\n

<------>

\n

echinopluteus

\n\n

but, not, for example,

\n

echinoderm larva pluteus early

\n

<------>

\n

echinopluteus

\n

besides being two (dotted) links away?

\n

I realize my original posting of the question may have been subject to misinterpretation, which is why I ask.  I know the difference between solid lines and dotted lines, but for my algorithm, any sort of link is important.  

\n

Thanks again.

\n

Best,

\n

J. Q.

", "rawMarkdown": "", "votes": null }, { "id": "62375", "postDate": "01/06/2015 18:52:12", "content": "

Hi J.Q.,

\n

The order in which I listed those echinoderm groups is deliberate. The dotted line relationships are (purposely) vague because it's also your job to figure out which relationships are strongest based on the images we gave you.

\n

We don't want to a priori dictate which groups have to be related to each other because 1) that could bias the classification scheme, 2) there are obviously feature-related connections between groups that are taxonomically distinct from each other that are not marked with dotted or solid lines, and 3) forcing relationships into the classification scheme make it less flexible in case groups are dropped, or added. 

\n

The plankton relationships guide is meant for you to use as a resource, a verification mechanism to make sure your program is bringing the right groups together. I would recommend using it at the data exploration phase and nothing further. But obviously, anything we give you is fair game. 

\n\n

cheers, and good luck,

\n

Jessica

", "rawMarkdown": "", "votes": null }, { "id": "62376", "postDate": "01/06/2015 19:06:41", "content": "

Hi Jessica,

\n

Of course, of course.  To be honest, I haven't bothered coding in any but the more obvious taxonomical and structural similarities.  There was one thing I wanted to know about relationships that were given a priori that I could verify a posteriori.  Ah well.

\n

Thanks!

\n

-J.Q.

", "rawMarkdown": "", "votes": null } ], "comments": [ { "id": 62372, "author_name": "jessicaluo", "author_url": "", "post_date": "01/06/2015 17:50:23", "content": "

Hi J.Q.,

\n

The dotted lines mean that there is some feature-related relationship between the groups. They're separate groups (not all alike), but similar to each other in some way. They are not derivative of each other in either taxonomy or shape, or there would be solid lines between them. 

\n

Hope this helps,

\n

Jessica

", "votes": null, "replies": [] }, { "id": 62374, "author_name": "jqveenstra", "author_url": "", "post_date": "01/06/2015 18:20:41", "content": "

Hi Jessica,

\n

Thanks.  So, just to be clear, we have, for example, 

\n

echinoderm larva pluteus early

\n

<----->

\n

echinoderm larva pluteus urchin

\n

<------>

\n

echinoderm larva pluteus star

\n

<------>

\n

echinopluteus

\n\n

but, not, for example,

\n

echinoderm larva pluteus early

\n

<------>

\n

echinopluteus

\n

besides being two (dotted) links away?

\n

I realize my original posting of the question may have been subject to misinterpretation, which is why I ask.  I know the difference between solid lines and dotted lines, but for my algorithm, any sort of link is important.  

\n

Thanks again.

\n

Best,

\n

J. Q.

", "votes": null, "replies": [] }, { "id": 62375, "author_name": "jessicaluo", "author_url": "", "post_date": "01/06/2015 18:52:12", "content": "

Hi J.Q.,

\n

The order in which I listed those echinoderm groups is deliberate. The dotted line relationships are (purposely) vague because it's also your job to figure out which relationships are strongest based on the images we gave you.

\n

We don't want to a priori dictate which groups have to be related to each other because 1) that could bias the classification scheme, 2) there are obviously feature-related connections between groups that are taxonomically distinct from each other that are not marked with dotted or solid lines, and 3) forcing relationships into the classification scheme make it less flexible in case groups are dropped, or added. 

\n

The plankton relationships guide is meant for you to use as a resource, a verification mechanism to make sure your program is bringing the right groups together. I would recommend using it at the data exploration phase and nothing further. But obviously, anything we give you is fair game. 

\n\n

cheers, and good luck,

\n

Jessica

", "votes": null, "replies": [] }, { "id": 62376, "author_name": "jqveenstra", "author_url": "", "post_date": "01/06/2015 19:06:41", "content": "

Hi Jessica,

\n

Of course, of course.  To be honest, I haven't bothered coding in any but the more obvious taxonomical and structural similarities.  There was one thing I wanted to know about relationships that were given a priori that I could verify a posteriori.  Ah well.

\n

Thanks!

\n

-J.Q.

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "62187": "", "62372": "", "62374": "", "62375": "", "62376": "" }, "source": "meta" }