MATLAB Community

MATLAB, community & more

Ask a Question in Answers with Auto-suggest 6

Posted by Ned Gulley,

This week’s post comes from Kent Millard, the UX Specialist for the MATLAB Answers team.

Quick Answers with Auto-Suggest

by Kent Millard

Let’s say you’re wondering how to plot some data contained in a 3D matrix. You’ve searched high and low to find an answer, but without satisfying results. Finally, you come across the MATLAB Answers page and there it is… the big blue “Ask a Question” button!


With gathering excitement, you press the button and begin to type.

This is where it gets interesting.


Before you even finish your query, similar questions that have already been answered start to appear. One of these might just have the information you need. If so, you get an instantaneous answer pulled from our database of more than 75,000 answered questions!

Here’s how it works.

Start typing your question—keywords or natural language work equally well. The most relevant, answered questions will appear below. The boxes with numbers tell you how many answers there are to a particular question. Green boxes mean the person asking the question confirmed that one of the answers solved their problem.

Questions with tiny blue icons (it’s a stylized MathWorks logo) have been posted by the MathWorks Support Team, and they always include an answer that solves the problem.

We think auto-suggest will save you time. But rather than read more about it, give it a try and tell us what you think!


Comments are closed.

6 CommentsOldest to Newest

Yair Altman replied on : 1 of 6

I took a spin at the new feature and I like the UI, but I think it could use a tune-up. The results do not appear to be sorted in a way that would be very relevant to readers so I’m afraid that if there are more than the handful of results for which there’s room in the small display box, users will not benefit much, and might actually be frustrated. Some examples: “performance” (only returns Simulink performance answers); “speed” (only few of the results are actually related).

To improve the ranking, I would suggest ordering the results based on the following criteria:

search term(s) [excluding stop words] appear in the thread title
search term(s) [excluding stop words] appear in the thread tags
search term(s) [excluding stop words] density within the thread contents
a thread having multiple answers is more important than one having only a few
a thread with multiple answers & comments is more important than a thread with less of them
a thread whose accepted answer was by a high-ranking member is more important than a thread with an answer by a lower-ranked member
a thread whose accepted answer had multiple up-votes is more important than a thread with an answer with less votes (or none, which is typical)
a thread whose accepted answer had multiple comments is more important than a thread with an answer having less (typically no) comments

I’m sure you can come up with plenty more criteria like these. Naturally you could play with the relative weights of these, but I think that any one of them could have solved the problems with “performance” and “speed” above.

The UI is nice, but if you don’t follow up with good content, it will all be for naught.

Zu replied on : 2 of 6

I am using the idnlgrey box tool. my model involves 8 ODEs with 18 unkown parameters. I have 10 experimental data for each of the 4 observable compartments (the variables). I dont know how to represent Ny in the idnlgrey box tool
Can you help please

Kent Millard replied on : 4 of 6

Excellent feedback, as always; thank you Yair.
Your suggestions are spot-on and in line with what we’ve talked about doing for auto-suggest 2.0. Many of the search criteria that you cited are currently being used within the feature, so we’re thinking that stop words might enhance the result set. As we develop the next iteration, we’ll reach out to you for your input.
Thanks again.

Sean replied on : 5 of 6

I’m trying to turn some bones from a CT scan into an iso surface… but it comes out looking like a bunch of Legos. What am I doing wrong? Is there any way to improve the surface, or am I fighting a loosing battle against slice-thickness resolution? My email is Im using: imstack… imfill(bw,’holes’)… centroid and area regionprops… thresh_tool… isosurface.