Beginner Questions involving Parsing

[Scott Hamilton]

I am a beginner with FLEx, so I hope you will excuse me if I ask some really basic questions. I have a few small questions about the parsers, and then a snag I have hit.

In terms of the differences between the parser, am I correct in my understanding that if you want to use the Phonological Rules under Grammar, you need to use the HermitCrab parser?

I read somewhere on here that it relies primarily on phonological features for parsing. But what about parsing that relies on things like inflectional features? For example, my understanding is that Affix process rules are primarily for the HermitCrab parser. However, if I convert an affix allomorph to a affix process rule, I lose the option for assigning a required inflection for that allomorph. Am I missing something about that?

Finally, I have an odd problem, but I am not sure if I am configuring FLEx properly. 

I have an affix that is a circumfix (i- -i) that is applied to nouns. The HermitCrab parser can parse that (it parses /ilakemi/ properly to i- -i + lakem). I also have an suffix (-n) which can be parsed properly (/mudin/ parses properly to mudi + n). However, the parser fails when the two are used together (/ilakemin/ fails parsing).

The circumfix is configured with an affix process rule: [X] -> i + 1 + i

The suffix is configured just as a regular suffix of the form: n

Looking at the steps, it parses down to lakem as the basic root, but fails when it tries to reform it. Instead of applying the circumfix again, it seems to want to build it as i + lakem + n (seems to miss the second half of the circumfix). How should I be configuring the affixes for this to work?

Thank you for your time

Scott

 

[Kevin Warfel]

Hi Scott,

 

A couple of quick remarks before I leave for the day. Someone else can complete the picture or I will try to address the other questions in the next day or two. (Comments inserted below.)

 

Kevin

 

From: flex...@googlegroups.com [mailto:flex...@googlegroups.com] On Behalf Of Scott Hamilton
Sent: Wednesday, November 11, 2015 4:13 PM
To: FLEx list <flex...@googlegroups.com>
Subject: [FLEx] Beginner Questions involving Parsing

 

I am a beginner with FLEx, so I hope you will excuse me if I ask some really basic questions. I have a few small questions about the parsers, and then a snag I have hit.

 

In terms of the differences between the parser, am I correct in my understanding that if you want to use the Phonological Rules under Grammar, you need to use the HermitCrab parser? It is true that the other parser ignores the Phonological Rules section, but there are other ways to apply “rules” if you use that parser.

 

I read somewhere on here that it relies primarily on phonological features for parsing. True. But what about parsing that relies on things like inflectional features? HC uses both. For example, my understanding is that Affix process rules are primarily for the HermitCrab parser. Affix Process Rules cannot be used with the other parser, but HC does not require APRs. However, if I convert an affix allomorph to a affix process rule, I lose the option for assigning a required inflection for that allomorph. It’s not obvious to me that this should be true. I probably need more details about what you mean in order to understand. Am I missing something about that?

 

Finally, I have an odd problem, but I am not sure if I am configuring FLEx properly. 

 

I have an affix that is a circumfix (i- -i) that is applied to nouns. The HermitCrab parser can parse that (it parses /ilakemi/ properly to i- -i + lakem). I also have an suffix (-n) which can be parsed properly (/mudin/ parses properly to mudi + n). However, the parser fails when the two are used together (/ilakemin/ fails parsing).

 

The circumfix is configured with an affix process rule: [X] -> i + 1 + i  I’m not sure why you chose do use an APR here. Can you not declare it as a simple circumfix? It’s been a while since I worked with something like this, so I’m not sure, but it’s not immediately obvious to me why you need an APR.

The suffix is configured just as a regular suffix of the form: n

 

Looking at the steps, it parses down to lakem as the basic root, but fails when it tries to reform it. Instead of applying the circumfix again, it seems to want to build it as i + lakem + n (seems to miss the second half of the circumfix). How should I be configuring the affixes for this to work? This might be related to the fact that you used an APR, but I suspect that it has do rather with your template(s).

 

Thank you for your time

Scott

 

--
You are subscribed to the publicly accessible group "FLEx list".
Only members can post but anyone can view messages on the website.
---
You received this message because you are subscribed to the Google Groups "FLEx list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to flex-list+...@googlegroups.com.
To post to this group, send email to flex...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/b4bd6c51-3710-4eee-8551-f29900ff9904%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Scott Hamilton]

A couple of quick remarks before I leave for the day. Someone else can complete the picture or I will try to address the other questions in the next day or two. (Comments inserted below.) 

Thank you for your time!

In terms of the differences between the parser, am I correct in my understanding that if you want to use the Phonological Rules under Grammar, you need to use the HermitCrab parser? It is true that the other parser ignores the Phonological Rules section, but there are other ways to apply “rules” if you use that parser.

Are there example files one can obtain to see some of these processes? I learn well from examples.

 

I read somewhere on here that it relies primarily on phonological features for parsing. True. But what about parsing that relies on things like inflectional features? HC uses both. For example, my understanding is that Affix process rules are primarily for the HermitCrab parser. Affix Process Rules cannot be used with the other parser, but HC does not require APRs. However, if I convert an affix allomorph to a affix process rule, I lose the option for assigning a required inflection for that allomorph. It’s not obvious to me that this should be true. I probably need more details about what you mean in order to understand. Am I missing something about that?

When you specific an allomorph for an affix, you can put inflectional features in the Required Featured field (as in this allomorph is only used on words of a certain noun class).

Let's say I convert an allomorph on an affix to an APR. The field for Required Features disappears. The field for Inflection Classes remains, but not for Inflection Features. So if I do need/want to use an APR, I can't choose which APR based off of a feature (unless I make several different headword entries, I think).

 

The circumfix is configured with an affix process rule: [X] -> i + 1 + i  I’m not sure why you chose do use an APR here. Can you not declare it as a simple circumfix? It’s been a while since I worked with something like this, so I’m not sure, but it’s not immediately obvious to me why you need an APR.

The suffix is configured just as a regular suffix of the form: n

This may simply be a result of my inexperience. I found an example in the help file about creating a circumfix and it mentioned using the APR to do so.

  

Looking at the steps, it parses down to lakem as the basic root, but fails when it tries to reform it. Instead of applying the circumfix again, it seems to want to build it as i + lakem + n (seems to miss the second half of the circumfix). How should I be configuring the affixes for this to work? This might be related to the fact that you used an APR, but I suspect that it has do rather with your template(s)

Actually, you are dead on about this. I looked at my noun template and looked at the discussions of using them with triconsonantal roots and realized I only had one slot in my template for my circumfix where I needed two. Thanks for the pointer!

Scott

[Andy Black]

On 11/11/2015 3:39 PM, Scott Hamilton wrote:

In terms of the differences between the parser, am I correct in my understanding that if you want to use the Phonological Rules under Grammar, you need to use the HermitCrab parser? It is true that the other parser ignores the Phonological Rules section, but there are other ways to apply “rules” if you use that parser.

Are there example files one can obtain to see some of these processes? I learn well from examples.

In a separate message, I can send you the Bahasa Indonesia project containing material mentioned in Appendix B of the conceptual intro document.

 

I read somewhere on here that it relies primarily on phonological features for parsing. True. But what about parsing that relies on things like inflectional features? HC uses both. For example, my understanding is that Affix process rules are primarily for the HermitCrab parser. Affix Process Rules cannot be used with the other parser, but HC does not require APRs. However, if I convert an affix allomorph to a affix process rule, I lose the option for assigning a required inflection for that allomorph. It’s not obvious to me that this should be true. I probably need more details about what you mean in order to understand. Am I missing something about that?

When you specific an allomorph for an affix, you can put inflectional features in the Required Featured field (as in this allomorph is only used on words of a certain noun class).

Let's say I convert an allomorph on an affix to an APR. The field for Required Features disappears. The field for Inflection Classes remains, but not for Inflection Features. So if I do need/want to use an APR, I can't choose which APR based off of a feature (unless I make several different headword entries, I think).

You are correct.  Affix process rules do not allow for inflection classes (and I'm not sure why...).

--Andy

[Scott Hamilton]

On Thu, Nov 12, 2015 at 9:11 AM, Andy Black <andy_...@sil.org> wrote:
> Are there example files one can obtain to see some of these processes? I
> learn well from examples.
> In a separate message, I can send you the Bahasa Indonesia project
> containing material mentioned in Appendix B of the conceptual intro
> document.

Thank you very much!

> When you specific an allomorph for an affix, you can put inflectional
> features in the Required Featured field (as in this allomorph is only used
> on words of a certain noun class).
>
> Let's say I convert an allomorph on an affix to an APR. The field for
> Required Features disappears. The field for Inflection Classes remains, but
> not for Inflection Features. So if I do need/want to use an APR, I can't
> choose which APR based off of a feature (unless I make several different
> headword entries, I think).
>
>
> You are correct. Affix process rules do not allow for inflection classes
> (and I'm not sure why...).

Ah, good to know I am at least not misunderstanding the application. I
wonder if there is a technical reason, or if it is just an oversight.

Anyway, thanks again!

Scott

[Andy Black]

On 11/12/2015 11:00 AM, Scott Hamilton wrote:
> On Thu, Nov 12, 2015 at 9:11 AM, Andy Black <andy_...@sil.org> wrote:
>> When you specific an allomorph for an affix, you can put inflectional
>> features in the Required Featured field (as in this allomorph is only used
>> on words of a certain noun class).
>>
>> Let's say I convert an allomorph on an affix to an APR. The field for
>> Required Features disappears. The field for Inflection Classes remains, but
>> not for Inflection Features. So if I do need/want to use an APR, I can't
>> choose which APR based off of a feature (unless I make several different
>> headword entries, I think).
>>
>>
>> You are correct. Affix process rules do not allow for inflection classes
>> (and I'm not sure why...).

Oops, I typed the wrong thing. I meant to say that Affix process rules
do not allow for "required features" (i.e., infection *features* which
must be present). If the affix is marked as inflectional, then it can
have inflection classes.

--Andy