search ordering and pressing ENTER for page result

[vsoch]

I think the ordering of the search might be organized in some fashion that (might be) more akin to what the user is looking for, although I can't tell you exactly what is right. But I can provide an example.

Just now, I did a search for "working memory." And probably the first result, as the "strongest" match should have been working memory (concept). But in fact the order was:

• Baddeley's model of working memory (theory)
• NCANDA Accuracy - Working Memory (battery)
• NCANDA Speed - Working Memory (battery)
• NIH Toolbox List Sorting Working Memory Test (task)
• Numerical Working Memory Task (task)
• RDoC Working Memory Matrix (theory)
• auditory working memory (concept)
• face working memory task (task)
• object working memory task (task)
• phonological working memory (concept)
• semantic working memory (concept)
• spatial working memory (concept)
• spatial working memory localizer task (task)
• patial working memory task (task)
• tactile working memory (concept)
• verbal working memory task (task)
• visual working memory (concept)
• working memory (concept)
• working memory fMRI task paradigm (task)
• working memory maintenance (concept)
• working memory retrieval (concept)
• working memory storage (concept)
• working memory updating (concept)

Notice that the "best" match was super far down the list! I think it's just returning based on alphabetical order, but grouped by something else first? I'm wondering if there is any way to put better logic into the ordering.

The second issue would be that I should be able to type "working memory" and then press enter to get an organized search summary view. That doesn't work right now, so if I am a first year graduate student coming to this interface, it's a bit hard or frustrating to use.

[poldrack]

+1

…

On Sun, Aug 19, 2018 at 10:13 AM Vanessa Sochat ***@***.***> wrote: I think the ordering of the search might be organized in some fashion that (might be) more akin to what the user is looking for, although I can't tell you exactly what is right. But I can provide an example. Just now, I did a search for "working memory." And probably the first result, as the "strongest" match should have been working memory (concept). But in fact the order was: - Baddeley's model of working memory (theory) - NCANDA Accuracy - Working Memory (battery) - NCANDA Speed - Working Memory (battery) - NIH Toolbox List Sorting Working Memory Test (task) - Numerical Working Memory Task (task) - RDoC Working Memory Matrix (theory) - auditory working memory (concept) - face working memory task (task) - object working memory task (task) - phonological working memory (concept) - semantic working memory (concept) - spatial working memory (concept) - spatial working memory localizer task (task) - patial working memory task (task) - tactile working memory (concept) - verbal working memory task (task) - visual working memory (concept) - *working memory (concept)* - working memory fMRI task paradigm (task) - working memory maintenance (concept) - working memory retrieval (concept) - working memory storage (concept) - working memory updating (concept) Notice that the "best" match was super far down the list! I think it's just returning based on alphabetical order, but grouped by something else first? I'm wondering if there is any way to put better logic into the ordering. The second issue would be that I should be able to type "working memory" and then press enter to get an organized search summary view. That doesn't work right now, so if I am a first year graduate student coming to this interface, it's a bit hard or frustrating to use. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#126>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA1KkCmIyZVyMGLmTz7UU-rGqRSneOxfks5uSZy-gaJpZM4WDCvY> .

-- Russell A. Poldrack Albert Ray Lang Professor of Psychology Professor (by courtesy) of Computer Science Bldg. 420, Jordan Hall Stanford University Stanford, CA 94305 [email protected] http://www.poldracklab.org/