Re: Re: Priority tagged Bliss lexicon - take 1

+1

Great work Mats and Galvin. Thank you for your collaboration. If you
require any assistance or need some extra hands on this feel free to reach
out.

Best,
Thaddeus

On Fri, May 4, 2018 at 1:46 AM, lisa.seeman <lisa.seeman@zoho.com> wrote:

> Hi Folks
>
> Please take a look at the wonderful work of Mats Lundalv to enable the
> symbols.  Also Russell Galvin from Bliss Symbolics has worked hard getting
> the URIs enabled.
>
> Note that this work is being done in coordination with the Unisef group
> for AAC AT,  which is a great basis for support in the future.
>
> I plan to add it to the personlization wiki
>
> All the best ... Lisa
>
> Dear all,
>
>
> Now I have a first version of a Bliss lexicon with priority tagging,
> matched against a number of English (mostly US) core vocabulary and word
> frequency lists.
>
> Note that this tagging is so far quite preliminary, subjective and not
> very "scientific". Hopefully still useful as a start.
>
>    - Most of the *644 highest priority 1* tagged Bliss-words come from
>    the Swedish-Scandinavian "Standard" template communication Bliss chart
>    vocabulary. This is so because that happened to be my starting point. It
>    contains a substantial part of the core vocabulary, but some of the
>    contained concepts are far from highest frequency in general, e.g. some
>    things regarded as specifically relevant for communicative needs of
>    children and people with profound disabilities. I have so far changed just
>    a few of them to prio 2, and this should be revisited.
>    - Note that the prio 1 group also includes the basic set of A-Z
>    letters, 0-9 numerical characters, 5 punctuation characters, and 6 Bliss
>    indicator characters, in total 48 entries of which most are not part of any
>    typical "core vocabulary".
>    - I have also so far added a few high frequency words/Bliss-words,
>    missing from this initial Bliss vocabulary, to the prio 1 group. It is of
>    course important to identify any additional words/concepts that should be
>    included there.
>    - The *prio 2 group* so far contains *127 words/Bliss-words*
>    - The *prio 3 group* so far contains *141 entries*
>    - The *prio 4 group* so far contains* 95 entries*
>    - The *prio 5 group* so far only contains *13 entries*, mostly because
>    I started adding prio 5 tags late in the process
>    - *Altogether 1020 Bliss-words are so long tagged 1-5* (including the
>    Latin alphabet characters). All non-tagged Bliss-words in the lexicon could
>    so far be regarded as priority 5 and beyond. A question is: How many
>    priority levels will be useful? (I definitely think that at least one or
>    two levels more could be useful to sort out outdated and very special parts
>    of the Bliss lexicon. (This may also go for some other symbol libraries -
>    like all the special Spanish region and places names and flags in ARASAAC,
>    etc.)
>
> Further work to be done:
>
>    - This priority tagging should ideally be further discussed, updated
>    and amended based on some more well thought through and defined criteria.
>    - I will go through the current list to sum up the actual number of
>    covered core vocabulary and high frequency English words that are actually
>    covered in the tagged part so far, as this is not the same as the number of
>    Bliss-words/concepts.
>    - I will also collate the list with such words that are not currently
>    covered in Bliss.
>    - Next step will be to compare this list with the VOA Special English
>    Word Book and the Wikipedia Basic English combined wordlist - with  a first
>    ambition to get an idea of how much of these lists that are already covered
>    among the priority tagged Bliss-words so far, or among the so far
>    non-tagged part of the Bliss lexicon, and thus also how much that is not
>    covered.
>
> You will find the tagged lexicon spreadsheet file at:
> http://www.symbolnet.org/bliss/BCI-AV_2018-04-16_(en+
> sv+no+fi+hu+de+nl+af+ru+lv+po+fr+es+pt+it+dk)+derivations+tag-sorted.zip
>
> You will find the work sheet that I have been using to match the material
> at:
> http://www.symbolnet.org/bliss/bliss_template+1st100+
> YACR-freq1000-329wl.ods
> This sheet includes comments for missing Bliss-words (rows with orange
> background colour). There you will also note that rows for words that are
> inflections of words covered by Bliss-words are marked with light yellow
> background colour.
> Bliss-word entries with green text colour are duplicates of entries
> elsewhere in the list.
>
> All comments and suggestions are welcome!
>
> Cheers,
> Mats
>
>
>
>
>
>
>
>
>