Equidox Webinar Wednesdays – Smart List Detector

See our new smart List Detector in action in this live demonstration. Watch as it quickly tags lists, even nested lists, in a few clicks. Learn how to get the most from this fantastic feature that saves you hours of remediation. Get your questions answered about this new artificial intelligence-powered feature.

Smart List Detector Demonstration

Tagging list items properly in a document allows the reader to understand how the content relates and that it is a unit.  This is done in one step with Equidox software, even for nested lists.

Video transcript

Transcript:

Equidox by Onix

Reach everyone.

Thank you everyone who is currently on the call. Thank you for joining

My name is Dan Tuleta. I’m a sales engineer for Equidox.

I may have met several of you in the past either events or on training calls

But the reason we called this webinar today was to take a little bit of a deeper dive

Into our new list detection technology that we’ve recently added to the to the platform

So there might be a mix of existing customers and users on this call or potentially people that are interested

to learn more about Equidox

But it occurred to us that many people might not have seen the new list detection tool in action

since your previous demos or trainings that you may have seen.

So we wanted to make sure that we are

Are supporting all of our users and making sure that everyone understands how to use the new tool correctly.

So with that said I’m going to jump into this document here, which has some different examples that we can take a look at.

And I believe that Tammy will be collecting. some questions. We’re going to try to get to questions if we have time

There are going to be many people on this call

so it might be difficult to answer every single one

What we do ask though is if you just type it into the Q&A part of the ZOOM platform there.

We will consolidate all these questions and potentially put out like a frequently asked questions document

to just make sure that every question is being answered.

Also with that said,

we will be available for follow-up questions and discussions and if you need

to schedule a quick call with me just to ask a specific question or go over anything

That you’ve been experiencing in using the new List Detection Tool

Please feel free to reach out to us at EquidoxSales@Onixnet.com.

Or, many of you have the direct email to either an account manager

or you can submit questions through our website as well.

Okay. So with that said,

We are now in the Equidox platform

And Ii’m looking here at a three-page document that just contains several lists for us to work on.

So i’m going to start here on this first page and we are just going to take a look

at the uh, at some simple stuff first just to make sure that everyone understands the basic functionality of the new list detection tool

so

Uh, on this page, we have a few very simple lists here. Just three standard list items marked with bullet points.

So with the old list technology, you might remember different text zones layered inside of the list zone.

Sort of tedious and clunky to deal with, still not very difficult to use.

We had a big blue detect button to automatically detect those items, but we’re doing things a little bit differently now.

So what we’re doing is still marking the individual zone that is wrapped around the entire list. We’re still marking that as a list zone

By changing the zone type in the drop-down menu

We still have that green bar that many of you might remember from the old list detection tool

And if we just place that list, that green bar, on top of the bullet points or whatever the list labels may be…

They might be dashes. They might be numbers. They might be letters…

We just place that green bar on top and then we have a new list detection slider down here.

So underneath the list properties we have the List Detector and if we just move this.

Uh, this blue dot on the slider from left to right one position.

You’ll see that that List Zone has now been broken up into three separate parts.

So when I take a look at the HTML preview of this page…

We will see that I have three separate list items that make up the entirety of that list.

The other zones have not yet been tagged as lists.

So that’s why they’re sort of rendering in this way where they’re just uh wrapped up in a text zone.

And they all just kind of turn into a strange run-on sentence. So we’re gonna have to touch those up as well

So that’s how we would deal with a very simple standard list.

Another thing to keep in mind when you’re dealing with lists…

If you for whatever reason cannot get the list to detect properly

You have the ability to set the amount of list items you would like in that list

manually. So there’s a little text field down here which contains a number based on how many list items the list detector was able to

find, you’re still able to change the amount of list items

by using the arrow keys or typing in the manual the number manually that you would like to add.

So if there is a sort of a list that’s not behaving for whatever reason, you can still

force the correct amount of list items and then you can grab the

list bars (these little divider lines here) and you can move them up and down accordingly to make sure that the data is placed

inside the individual list item.

Another great thing about this new List Detection tool is the ability to grab the entire list zone and move it around the page

as needed. If you remember, before we had all of the text zones wrapped inside of the list zone and as you moved one zone

nothing else would move with it.

So now if you have an instance where your list is very tightly spaced and perhaps the

the actual list items are just being uh…

The list detection lines are just a little bit too high or a little bit too low,

you can now just grab the entire list and move it down to cover

the list appropriately so that all of your items are making it in between the corresponding divider lines.

So that’s a nice feature. You’re not having to deal with so many zones anymore. You’re not having all those layering issues. You simply just

have these divider lines which can be adjusted and moved around the page.

So that’s a very simple three-item list just using bullet points.

We have another list here. And this is a list using numbers for the list labels.

And I get, I get a lot of questions about ordered lists.

And what the difference is in terms of how do we remediate them?

And how does it behave in a PDF document?

This is a question that comes up all the time.

So if you were ever wondering about what the difference between an ordered or an unordered list is, don’t feel bad.

Everyone asks all the time. So it’s the same process where we hit “L” on our keyboard

I’ll remind you of the keyboard shortcut to set the zone type to list.

By just tapping “L” while you have that zone selected and we have the green bar again that comes down the left-hand side.

If I just place that on top of the numbers and I use the List Detector again, just moving it over one slot…

It looks like I’ve moved the green bar too far to the right. So I’m going to try that again.

Okay, that worked perfectly. I had the

the green bar was just missing the number slightly.

So I had to just make a slight adjustment there.

And then trying it again, was able to fix that correct or was able to make that correction.

So

Everything here is just the same as what we’ve done up here. So when I look at the preview…

What we see, is currently there are three bullet points and then the actual numbers are included in the individual

list item.

Now this is some this is an aesthetic of HTML the way that HTML renders this data.

It’s something I’m asked about on a regular basis…

So keep in mind when you’re remediating lists if you are ultimately going to export the document as a PDF,

you do not know you do not have to worry about this next step.

By setting the list to an ordered list rather than the default setting (which is just an unordered list)

so if you’re exporting it to PDF,

everything else that I’m saying here is really just an extra step that has no… ultimately no payoff for you.

So you’re just doing extra work. It’s not going to change anything about the exported PDF.

But if you want to change it to an ordered list…

What you can do is if you just tap the radio button here for ordered…

That’s going to give you a drop-down that will appear below and this is very similar to the old List Detector.

For anyone that remembers the the old technology…

When I hit the drop-down menu, and I change it to numbers that is going to change the style type.

Only as it’s represented in the HTML preview and if we were to export this document at has HTML.

So when I look at the preview again…

Here I have

preset list labels, which are one, two, and three, and then I still have the one, two, and three inside of the list item.

So we have this duplicated look which is sort of odd to look at. And again, it’s something that I’m asked about all the time

So if if you are struggling with this I’m gonna show you right now the way that we can resolve this issue.

So, part of the problem is is that we’re just not detecting these numbers as the list labels.

So what we can do to make sure that we’re

absolutely going to detect these as the list labels is we’re going to choose the…

We’re going to select the list and then at the bottom here underneath List Detector. We’re going to press the show labels button.

That’s a little check box there.

So if I select that I can then individually select the individual list item and I can choose to turn on the list label.

Turn it on or off into the individual list items.

so

if I…

If I just double-click inside of these zones, I can turn on the list labels.

And then you’ll see that there’s now if I zoom in slightly more you’ll see that there’s these sort of checkered outline boxes

So this is where Equidox is looking for that list label. We can move these around.

We can resize them using the height and width.

So if I shrink down these actual list labels to the appropriate size.

And just make sure that they’re layered on top of the actual numbers themselves…

That’s going to help Equidox. Just determine that those numbers are in fact the list labels. I mean it does not need to then apply

Its own sort of HTML list labels that we are used to seeing.

So if I make those quick corrections, and if I just refresh the preview

you’ll see now I don’t have that duplicated number. I just have the one, two, and three and then straight into the actual list item.

So that’s how you would correct those types of errors where you might have a list that wasn’t actually created using list styles.

Maybe someone just typed the letter “a” with a period and started typing after that…

And created a very long list that way.

If you’re not able to detect those list labels,

you can still manually locate them using that workflow of turning on the list label button.

And then individually selecting the individual items by double-clicking inside of them and then turning on the list label

for that specific item and then just making sure that it’s layered on top.

So that’s a question that I’m frequently asked and I hope that clears up some questions

ahead of time if anyone was ever wondering about that.

We hope to avoid those situations, whenever possible.

And like I said before, this is only for instances where you’re exporting to HTML if you’re exporting to PDF

None of this is going to have an impact on that exported PDF file.

You can simply just use the Detection tool and even if you see the duplicated list labels on the HTML preview.

You can just ignore it when it gets exported a PDF. Everything is going to come out correctly.

Okay, so

the real reason we’re here today is to talk more about nested lists because nested lists are much more complex.

Or they used to be at least…

Because of the just the tedious nature of them of having to identify individual lists inside of list items

which make up a grander list so with our new technology which leverages

machine learning and computer vision.

We have a much better way of detecting nested lists automatically.

So our data scientists have been working very hard training models on how to identify

where these nests are and how do we place them inside of the parent list item, and then obviously grouping everything together as one

grand list.

So what we do now to detect lists

much much easier than ever before… if I just tap “L” on my keyboard to change the zone type to list…

If I move the List Detector of the slider… if I move this over…

Maybe I’ll try a setting of two in this case. You’ll see that it’s automatically identified that the main list items.

There’s two main items.

The top and bottom, which are broken up by this divider bar that you can see and then each one of those list items.

You have a separate list which are each broken into three separate list items.

So all of that happened by simply moving the List Detector from left to right.

Just two slots and it takes just a couple of seconds.

And you’ll see that we have a perfectly generated HTML list.

And we have not had to

do any sort of layering of different zones. We’ve not had to build it from the inside out

We simply place a zone over the entire list,

and then we press the… or we use the List Detector to do all of that heavy lifting for us.

So that’s how easy it is now in Equidox to work with nested lists.

You know, dramatic change to what we were doing before.

And if you’ve ever done this manually in a tool like perhaps Adobe Acrobat,

you’ll know that nested lists are just very tedious and they take a long time…

and a lot of patience.

So that’s a pretty simple page with just a couple of list examples i’m going to move over to the next page.

Where we actually have a more a more complicated nested list.

We have a much larger list. There’s tons of list items. We’ve got about three layers of nesting in this this is something that’s more

complicated but, you know,

if you were doing this the old way or if you’re trying to do this manually you could easily spend,

you know, a half an hour doing this in Adobe Acrobat, but with Equidox, all we’re doing is the same thing.

We’re pressing”L” and I’m going to move that List Detector over about four slots.

So what we’re doing is just changing sort of the algorithm and how focused it’s detecting the data.

So the more you move it over that’s typically going to give you a better result for these

more densely and more complicated nested lists. So you can choose this setting (there’s five separate settings one, two, three

four and five) you can kind of experiment with them and figure out what’s going to work for your specific nested list.

But in this case, I believe four is correct. And when I take a look at the actual

list that it’s created automatically here. Everything looks nice and clean. So

I’ve literally moved that from left to right, it took a couple of seconds. When I press the preview button

we have a perfect nested list automatically created for us and I haven’t had to touch any individual item.

I haven’t had to touch an individual label.

I’ve simply moved the Detector from left to right.

So very simple stuff there. You know, nested lists, previously were one of the more tedious… complicated…

Something that, you know, a new user might have to ask for help on.

But now you can see that just about anyone can be trained to handle nested lists in Equidox.

It’s just a matter of moving that Fetector from left to right and Equidox does it all for you.

Okay, so

I want to take a look here at the next page which has a few more examples on it.

That might get a little bit more complicated.

So, list merges are something that we are often asked about as well. So

if you have an instance where, this could be any number of reasons why this would occur, but you have a list

and then the actual list is disrupted by something (perhaps it’s a break in the column)…

Maybe it picks up on the next column. Maybe it’s a page break…

so the list ends at one page and continues on to the next…

Maybe there’s an image inserted in between two lists and you have to sort of merge the list together around them.

So the way that we do this now is just a little bit different our list… or our merge feature.

that we have added recently in our most recent major major build

has some new functionality to it that really helps

to sort of work around

PDFs that contain a lot of barriers where… whether it be images or page breaks or whatever…

It might be you can still merge to a target zone.

Um, so to make it a lot easier now instead of having to, you know, make any sort of post export touch-ups…

So now what we’re doing is we’re

still remediating these lists as normal.

So I’m just going to tap the “L” on my keyboard and I’m just going to move it over slightly and

move the Detector from left to right.

And my internet might be a little slow… there we go.

And I’ll do the same thing here where I’m just going to move the Detector from left to right…

And that looks pretty good. So I have two lists created with three items. When I look at the preview,

you’ll see I have my top set of list items and I have my continuation of that list down here.

But it’s being disrupted by that image. Well, I don’t want that to happen.

So I want to make sure that these are going to be grouped together as two, or as a single list.

Even though we have two separate lists here visually appearing on the page.

So what I’ll do is I’ll select the second list here and I will now turn on the merge button.

So I’ll press that merge checkbox right there.

Which is available on most of the zone types for text and headings and things like that.

So once you press that merge checkbox,

a new feature will appear underneath. So once you’ve checked merge, it’ll now ask you “with

what?” and then you can choose a list zone on this page.

And when you choose that a new drop-down will appear. And it’s now the zone…

And this drop-down will populate a list of feasible merge candidates.

So since I only have one other list on this page my drop-down menu here is just a

single zone that I can select. So when I select that zone

It’s going to merge this zone with that zone since this is the target zone that I’ve identified during that merge selection. process

So when I preview this…

You will see that I now have a single list that has been joined together through that merge.

And then the image will fall in the reading order after that list.

So that is how we are now merging lists together and this allows you to merge

to any target zone. So if your list

Is maybe it’s a magazine article and you have a list…

and then you have two pages of just advertisements or stock images and then two or three pages later your list continues on

You can

create those two lists separately, and then you can merge the two together.

Even across multiple pages. So the way that that would work,

just for example’s sake, is if you choose to merge to a list zone on-page and then a little dot dot dot

It’ll then give you the… ask you the same question of page and then you have a drop-down menu. You can select

any number of any of the pages that contain lists.

So once you select a page, you’ll then have a drop-down menu, and these are all the lists that are available to merge to

on page one.

So that’s how you can merge to target zones.

Identifying which page it’s on

and which zone you would like to merge to and Equidox will do all of the hard work for you.

So, one other question that we are asked a lot is about when you see this drop-down menu

you’ll notice that some of the

the naming conventions are a little bit strange. List 23, list 13, page 1 zone 6…

These are just automatically generated names by Equidox either when the document was imported or when the zone type was changed.

So if you ever are curious about this, what you can do is, if you if you’d like to merge to a target zone…

And you want to make sure that you’re definitely merging to the correct one…

What you can do is, you can actually name that zone something different. So the zone name is up here at the very top,

So when you change the zone name, you can say, you know “merge to me.”

You can type in whatever you would like. This does not impact the exported file whatsoever.

This is simply for your internal sort of process while you’re remediating the document to help you make sure that you are

linking to the correct zone or merging to the correct zone, I should say.

So, when you hit the drop-down menu again…

If I go “merge to list zone” on this page when I hit the drop-down

You’ll see no longer do I have that strange

Name of that zone, It now says “merge to me.”

So you’re able to just identify it much more easily when you change the zone name of that specific zone you would like to merge

to by just typing in a new name for that zone at the very top while you have that

zone selected. So you can name that whatever you would like to just uniquely identify it, to make sure that,

you are in fact merging to the correct page, and the same can be said for text zones as well.

Keep that in mind, you know, not every element in a PDF is a list…

So if you have 50 different text zones on a page and you want to make sure that you’re merging to the correct one,

that’s a really helpful feature, to be able to change that zone name.

To make it obvious to you as the remediator that you are, in fact, merging to the correct zone.

So keep that in mind if you have a lot of list merges to handle, you can change that zone name.

That’s something that is…

Can be very very helpful.

Okay, so

again, sometimes we have lists without style. Meaning that someone didn’t actually create

a list with the…

with…

Using the list styling, like in Microsoft Word.

For example, if you choose from the top menu, the top ribbon, the bullet points or the numbering system…

Someone might literally just type in “a” or they might type a dash and then just start typing text after that…

Which can be sometimes difficult for list detection to actually pick up.

So this is just a reminder that if you ever run across those instances…

You can certainly just change the number of list items manually by typing in three or using the arrow buttons.

And then you can place these accordingly by dragging them around. So keep that in mind, occasionally you’ll have,

you know, lists can be tricky. Just depending on how they’re designed.

But you have plenty of tools at your fingertips with Equidox, whether it be the automated List Detector,

which will work for you in probably 99 of cases.

But, you know, occasionally you’ll run across a very poorly designed list…

Or someone was just kind of typing away and they were using little dash marks instead of actually using list styles.

If your detection is not perfect,

You can always just force that amount of list items into that specific list and then move around your list items…

the list divider bars accordingly.

Okay. So, we’re doing pretty well here. We’re gonna we got a couple more examples and we got a few minutes left.

So we’re right on track, I think.

And, so here’s a list with

three separate items. But these lists actually contain links, as you can see.

So when you have a list that contains links, that’s pretty common, I feel.

What you can do is, you don’t really have to do anything differently because these links were embedded in the document.

So Equidox is actually able to identify them as links.

You can see they all have their own individual zone placed on top of them. So when you use the List Detector…

If you move the bar over, typically a setting of two when you have a standard list that contains links, I found that the second…

The second setting is most accurate.

When you look at the preview using this method,

you will see that we have the list

items detected, and they also contain those links, which you can click on and navigate to.

If you’d like to just verify that they are in fact working..

You can also select the individual links and just double-check the anchor. But as I said before, if the link is embedded,

you’re not going to have that issue.

You know, typically the anchor point, assuming that it’s accurately…

was accurately embedded in the document,

everything will just be taken care of by just simply importing it. And then once you mark it as a list,

those anchors those URLs or whatever…

They might be… those links are going to stay active within uh enclosed in the individual list item.

Now if you have a list that you don’t have an embedded link in the document…

Let’s say for example, if you just, if we get rid of that link, for example…

If you want to still manually create a link…

Maybe it’s like a directory or something like that, where it’s a list of email addresses or something…

You can still turn those into actual email addresses.

Or URLs. Whatever it may be.

You simply just can manually create that link zone. And then you can type in the anchor point that you would like.

So if we just pretend that (although it’s google.com) if we want it to be an email address, for example…

if I type in “mailto”

so

M-a-i-l-t-o and then a colon and then I can type in an email address…

So if I just type in onixnet.com…

When I look at the preview

I will still have that that that text representing that it’s a google link, but when I click on it…

it’s actually going to open up my email and compose an email to someone. So

that’s how that can work. If you don’t have an embedded link, you can still place them inside of list items.

It’s just that extra step of creating the linked zone and manually typing in the link… snchor….

where you would like that link to be directed to.

So “mailto” is a common one.

If it’s an email address that you would like to open up… and then also if it’s a web address your typical HTTPS

Colon, slash, and then the web address to follow after that.

If you have a link to for example, like a table of contents, your anchor point should start with

hashtag, page, underscore.

Just keep that in mind when you’re dealing with lists that contain links.

Your anchor point and the appropriate prefix for that type of link.

Okay, so we have one other example here that I wanted to look at, and this is an example of how you can identify….

You can still utilize the custom zone source, which many of you may be familiar with if you’ve used Equidox for a while.

Here’s an example of where we have a glaring misspelling in the list item.

So since this is a PDF and this version is sort of locked. We’re not here to edit this visually on the..

um

on the…

on the PDF itself

But to make sure that a screen reader

is not going to inaccurately read this

to someone that might not be able to see that. This is clearly a glaring misspelled word. It says “costume” instead of “custom.”

What we can do is still tag it as a list as normal…

And then once I’ve done that I can double click inside the individual list items…. and if I go to the source…

I can choose “custom”

And I can type in

“custom…

list..

item one..”

Instead of it the glaring misspelling of “costume list item one.” So for anyone that’s using this interacting with this document with a screen reader…

Those types of misspellings can create a lot of confusion.

And you have the ability in Equidox to make sure that the screen reader is not going to

inaccurately convey a clearly misspelled word.

You can go in there and make that change and type it into the custom field and when you look at the preview.

You will see that instead of it saying “costume list item one” it now says “custom list item one” and then list item two

I have not yet changed..

But I can do that very quickly. Again, just by double-clicking inside the list item, selecting custom, and then going to typing in

“Custom list item two,” and then refreshing the preview.

I will see that I in fact have corrected that spelling now.

Like I said before, it’s not going to visually update the PDF. We didn’t generate this PDF.

We have just been asked to remediate it.

So that’s not really our fault that someone made a glaring misspelling on that PDF document.

But what we can certainly do is fix it within the tagged reading order to make sure that

individual list, and that individual list item, is going to be read correctly

through that screen reader.

So that’s how we can use the custom zone source

inside of the individual list items to make those types of corrections.

You have that flexibility to insert into this text field, whatever you would like

that individual list item to read. And again, you just double-click inside the individual list item to identify it…

You’ll see that it turns a shade of blue instead of pink and that lets you interact with that individual list item.

Okay, so, we are just about at the 2:30 mark.

Tammy, I don’t know if there’s anything that you wanted to finish up with or follow or if there was any,

you know, common question that was asked multiple times in the Q&A.

Tammy: Actually, um

Thank you, Dan.

We only had a couple of questions. One of which was was whether or not this was going to be recorded… and yes it is.

I will be sending that around to everyone who registered. It’ll probably take a day or two. I’ve got to get it

captioned… But as soon as that happens,

everyone will get a copy so you can review it or share it with anyone else that you work with that might be interested.

The only other question, you answered within the demonstration, so I think we are good.

But of course if anyone has any follow-up questions, you can certainly email them to us.

EquidoxSales@onixnet.com

or from the app.

Dan: Great thank you very much, and it’s right at 2:30, so

we will be respectful of everyone’s time…

And again just to reiterate, please feel free to reach out

if you have any individual questions or individual

use cases that you would like to discuss. I’m happy to work with you to

figure out the best workflow and make sure that you’re using theList Detector correctly.

So thank you very much, everyone. I’m going to stop sharing and we can end the Webinar. Tammy: I do have one more thing

Dan, I wanted to let everyone know that we will be

doing another webinar to cover the new

Table Detector

that’s just been rolled out. And that will be happening in a couple of weeks.

So all of you who are on this registration list

will be getting an invitation to that, as well.

Dan: Perfect. Thank you for reminding me of that. So, yep, looking forward to that next demonstration…

And yeah, we will we’ll see and chat with everyone soon. Thanks, everyone!

END OF TRANSCRIPT.

Tammy Albee

Tammy Albee | Director of Marketing | Equidox Tammy joined Equidox after four years of experience working at the National Federation of the Blind. She firmly maintains that accessibility is about reaching everyone, regardless of ability, and boosting your market share in the process. "Nobody should be barred from accessing information. It's what drives our modern society."