> Ahh, accidentally posted without continuing!!
>
> Here's the full post again:
>
> I'm building  a vb6 application with ado 2.81 and access 2003 as a backend.
> The app parse recipe files and import ingredients into a table.
>
> Now, let's take as an example 8 different recipes that have the same
> ingredient (eg. cilantro). When I parse these 8 recipes I will check if the
> ingredient is not already available in the table. If it is, i'll use it in
> the recipe. If not, I have to create a new one. So let's take these 8
> recipes where they have the following ingredient line for cilantro:
>
> Snipped fresh cilantro
> Cilantro -- finely chopped
> Leaves of Cilantro -- optional
> Chopped cilantro
> Cilantro, fresh, chopped
> Cilantro, chopped
> fresh cilantro leaves -- coarsely chopped
> Cilantro, minced
>
> The whole idea of the app is to have only one ingredient called cilantro and
> not 8. How would one go about designing an intelligent function that
> recognizes that cilantro is the ingredient and will use that instead of
> creating 8 different records in the table?

"Larry Lard" <larryl***@hotmail.com> schrieb im Newsbeitrag
news:1128687301.185126.39100@g43g2000cwa.googlegroups.com...
>
> Ivan Debono wrote:
> > Ahh, accidentally posted without continuing!!
> >
> > Here's the full post again:
> >
> > I'm building  a vb6 application with ado 2.81 and access 2003 as a
backend.
> > The app parse recipe files and import ingredients into a table.
> >
> > Now, let's take as an example 8 different recipes that have the same
> > ingredient (eg. cilantro). When I parse these 8 recipes I will check if
the
> > ingredient is not already available in the table. If it is, i'll use it
in
> > the recipe. If not, I have to create a new one. So let's take these 8
> > recipes where they have the following ingredient line for cilantro:
> >
> > Snipped fresh cilantro
> > Cilantro -- finely chopped
> > Leaves of Cilantro -- optional
> > Chopped cilantro
> > Cilantro, fresh, chopped
> > Cilantro, chopped
> > fresh cilantro leaves -- coarsely chopped
> > Cilantro, minced
> >
> > The whole idea of the app is to have only one ingredient called cilantro
and
> > not 8. How would one go about designing an intelligent function that
> > recognizes that cilantro is the ingredient and will use that instead of
> > creating 8 different records in the table?
>
> In general, a difficult problem. If you do end up creating something
> that can parse descriptions like this as well a person, be sure to tell
> us, as you would probably be eligible for an AI prize or two :)
>
> Depending on how many recipes you have to process, you should first
> consider the time it will take to develop an automatic procedure versus
> the time it would take to do the data entry manually. The latter would
> be long and tedious but easy. The former might end up getting you
> nowhere.
>
> Here are some thoughts on how to approach this:
>
> Make various lists that contain words that might be in the recipe, but
> are not ingredients themselves. eg:
>
> - a list of verbs (and their various inflections) that describe ways in
> which ingredients might be prepared. Thus: chop(ped), snip(ped),
> mince(d) etc etc
>
> - a list of adjectives that describe ingredients. Thus: fresh, boiled,
> raw, premium etc
>
> When presented with an input string, remove any words that match
> entries on these lists. What you end up with might be ingredients. Look
> at the the outputs you get an iteratively add things to your lists. For
> example, your sample inputs end up (after removing punctuation also) as
> :
>
> cilantro
> Cilantro finely
> Leaves of Cilantro optional
> cilantro
> Cilantro
> Cilantro
> cilantro leaves coarsely
> Cilantro
>
>
> So 'optional' becomes another list candidate, as do various adverbs
> that might modify the verbs you already have. Have a list of 'small
> words' eg a, an, the, of, in etc and your list becomes (after forcing
> to lower case) :
>
> cilantro
> cilantro
> leaves cilantro
> cilantro
> cilantro
> cilantro
> cilantro leaves
> cilantro
>
> We are almost there. 'Leaves' is tricky because it might actually be
> part of the ingredient (eg bay leaves). The final step would then be to
> have a mapping table that maps both 'cilantro leaves' and 'cilantro' to
> your actual ingredient record.
>
> You will find that this whole process will be a highly *iterative* one
> - set up your filters, run through the data, see what you get, refine
> your filters, run again, refine, run, ... until you end up with data
> that you can just put into a mapping table as described above, and then
> you are done.
>
> Until of course your next batch of input contains a word you've never
> encountered before...
>
> --
> Larry Lard
> Replies to group please
>

"J French" <erew***@nowhere.uk> schrieb im Newsbeitrag
news:4346645a.2849493@news.btopenworld.com...
> On Fri, 7 Oct 2005 12:55:22 +0200, "Ivan Debono"
> <ivanm***@hotmail.com> wrote:
>
> <snip>
>
> >Snipped fresh cilantro
> >Cilantro -- finely chopped
> >Leaves of Cilantro -- optional
> >Chopped cilantro
> >Cilantro, fresh, chopped
> >Cilantro, chopped
> >fresh cilantro leaves -- coarsely chopped
> >Cilantro, minced
>
> >The whole idea of the app is to have only one ingredient called cilantro
and
> >not 8. How would one go about designing an intelligent function that
> >recognizes that cilantro is the ingredient and will use that instead of
> >creating 8 different records in the table?
>
> That is very interesting
>
> I recommend that you create a Lexicon
> - ie: a sorted list of words
>
> But not of the ingredients, rather of words that are not ingredients
> - eg: chopped   '--'    optional   ','   of
>
> You might also make a list of adjectives, eg: Fresh, Dried
> - as there is a difference between Fresh Cilantro and Dried Cilantro
>
> Heck - you might as well have a list of ingredients as well
>
> By the time you have removed the 'non ingredients' what will be left
> will be an ingredient - and optionally an adjective
>
> The reasoning is that there are probably a lot more ingredients than
> non-ingrediants - also the system will 'fail safe' one can tolerate a
> bit of garbage, but missing a vital ingredient will screw thing up big
> time.
>
> Obviously this will require quite a lot of manual work ticking words
> off as non-ingredients, but a few simple utilities (written by you)
> will make that fairly simple.
>
> I can see a few problems with things like Vanilla Essence versus
> Vanilla Pod, but I think you'll get round that with the optional
> adjectives, failing that you could make compound words
>
> Incidentally quantities need looking at
>
> I've done similar things before as spelling checkers and search
> accelerators - they work quite nicely, especially when 'trained'
>
> I would just use text files for this.
>
>

"Ivan Debono" <ivanm***@hotmail.com> wrote in message
news:eB9hUzyyFHA.736@tk2msftngp13.phx.gbl...
> Ahh, accidentally posted without continuing!!
>
> Here's the full post again:
>
> I'm building  a vb6 application with ado 2.81 and access 2003 as a
backend.
> The app parse recipe files and import ingredients into a table.
>
> Now, let's take as an example 8 different recipes that have the same
> ingredient (eg. cilantro). When I parse these 8 recipes I will check if
the
> ingredient is not already available in the table. If it is, i'll use it in
> the recipe. If not, I have to create a new one. So let's take these 8
> recipes where they have the following ingredient line for cilantro:
>
> Snipped fresh cilantro
> Cilantro -- finely chopped
> Leaves of Cilantro -- optional
> Chopped cilantro
> Cilantro, fresh, chopped
> Cilantro, chopped
> fresh cilantro leaves -- coarsely chopped
> Cilantro, minced
>
> The whole idea of the app is to have only one ingredient called cilantro
and
> not 8. How would one go about designing an intelligent function that
> recognizes that cilantro is the ingredient and will use that instead of
> creating 8 different records in the table?
>
> Thanks,
> Ivan
>

"Ralph" <nt_consultin***@yahoo.com> wrote in message
news:orydnShpNqGM8NvenZ2dnUVZ_smdnZ2d@arkansas.net...
>
> "Ivan Debono" <ivanm***@hotmail.com> wrote in message
> news:eB9hUzyyFHA.736@tk2msftngp13.phx.gbl...
> > Ahh, accidentally posted without continuing!!
> >
> > Here's the full post again:
> >
> > I'm building  a vb6 application with ado 2.81 and access 2003 as a
> backend.
> > The app parse recipe files and import ingredients into a table.
> >
> > Now, let's take as an example 8 different recipes that have the same
> > ingredient (eg. cilantro). When I parse these 8 recipes I will check if
> the
> > ingredient is not already available in the table. If it is, i'll use it
in
> > the recipe. If not, I have to create a new one. So let's take these 8
> > recipes where they have the following ingredient line for cilantro:
> >
> > Snipped fresh cilantro
> > Cilantro -- finely chopped
> > Leaves of Cilantro -- optional
> > Chopped cilantro
> > Cilantro, fresh, chopped
> > Cilantro, chopped
> > fresh cilantro leaves -- coarsely chopped
> > Cilantro, minced
> >
> > The whole idea of the app is to have only one ingredient called cilantro
> and
> > not 8. How would one go about designing an intelligent function that
> > recognizes that cilantro is the ingredient and will use that instead of
> > creating 8 different records in the table?
> >
> > Thanks,
> > Ivan
> >
>
> Quick First Brush White Board
>
> // table_Recipe   DDL
> ID:            seq_auto_number
> Name:       Text
> ' DataView
> 1    Soup
> 2    Meatloaf
>
> // table_RecipeIngredientsXREF
> RecipeID:      FK (number)
> IngredID:        FK (number)
> ' dataview
> 1    2
> 1    3
>
> // table_Ingredient  DDL
> ID:             seq_auto_number
> Item:          FK (number)
> Source        FK (number)
> Preparation: FK (number)
> ' dataview
> 1    1     1    1
> 2    2     2    5
> 3    3     1    1
>
> // table_Item    DDL
> ID:        seq_auto_number
> Name:    Text
> ' DataView
> 1 Cilantro
> 2 Onion
> 3 Celery
>
> // table_Source   DDL
> ID:            seq_auto_number
> Source:       Text
> ' DataView
> 1    Fresh
> 2    Canned
> 3    Bagged
>
> // table_Preparation   DDL
> ID:            seq_auto_number
> Prepare:    Text
> ' DataView
> 1    chopped
> 2    corsley chopped
> 3    snipped
> 4    whole
> 5    minced
>
> Should give you an idea.
> Note: You may not want to use an auto_number
> A complete unique readable tag would do.
> There are pros and cons.
>
> hth
> -ralph
>

"Ralph" <nt_consultin***@yahoo.com> schrieb im Newsbeitrag
news:3JidnUb_G7Lw7dveRVn-hg@arkansas.net...
>
> "Ralph" <nt_consultin***@yahoo.com> wrote in message
> news:orydnShpNqGM8NvenZ2dnUVZ_smdnZ2d@arkansas.net...
> >
> > "Ivan Debono" <ivanm***@hotmail.com> wrote in message
> > news:eB9hUzyyFHA.736@tk2msftngp13.phx.gbl...
> > > Ahh, accidentally posted without continuing!!
> > >
> > > Here's the full post again:
> > >
> > > I'm building  a vb6 application with ado 2.81 and access 2003 as a
> > backend.
> > > The app parse recipe files and import ingredients into a table.
> > >
> > > Now, let's take as an example 8 different recipes that have the same
> > > ingredient (eg. cilantro). When I parse these 8 recipes I will check
if
> > the
> > > ingredient is not already available in the table. If it is, i'll use
it
> in
> > > the recipe. If not, I have to create a new one. So let's take these 8
> > > recipes where they have the following ingredient line for cilantro:
> > >
> > > Snipped fresh cilantro
> > > Cilantro -- finely chopped
> > > Leaves of Cilantro -- optional
> > > Chopped cilantro
> > > Cilantro, fresh, chopped
> > > Cilantro, chopped
> > > fresh cilantro leaves -- coarsely chopped
> > > Cilantro, minced
> > >
> > > The whole idea of the app is to have only one ingredient called
cilantro
> > and
> > > not 8. How would one go about designing an intelligent function that
> > > recognizes that cilantro is the ingredient and will use that instead
of
> > > creating 8 different records in the table?
> > >
> > > Thanks,
> > > Ivan
> > >
> >
> > Quick First Brush White Board
> >
> > // table_Recipe   DDL
> > ID:            seq_auto_number
> > Name:       Text
> > ' DataView
> > 1    Soup
> > 2    Meatloaf
> >
> > // table_RecipeIngredientsXREF
> > RecipeID:      FK (number)
> > IngredID:        FK (number)
> > ' dataview
> > 1    2
> > 1    3
> >
> > // table_Ingredient  DDL
> > ID:             seq_auto_number
> > Item:          FK (number)
> > Source        FK (number)
> > Preparation: FK (number)
> > ' dataview
> > 1    1     1    1
> > 2    2     2    5
> > 3    3     1    1
> >
> > // table_Item    DDL
> > ID:        seq_auto_number
> > Name:    Text
> > ' DataView
> > 1 Cilantro
> > 2 Onion
> > 3 Celery
> >
> > // table_Source   DDL
> > ID:            seq_auto_number
> > Source:       Text
> > ' DataView
> > 1    Fresh
> > 2    Canned
> > 3    Bagged
> >
> > // table_Preparation   DDL
> > ID:            seq_auto_number
> > Prepare:    Text
> > ' DataView
> > 1    chopped
> > 2    corsley chopped
> > 3    snipped
> > 4    whole
> > 5    minced
> >
> > Should give you an idea.
> > Note: You may not want to use an auto_number
> > A complete unique readable tag would do.
> > There are pros and cons.
> >
> > hth
> > -ralph
> >
>
> Forgot to add that the tables would serve as both storage for "lookup"
keys
> and also for the "gathering". A true expert system will also include
> 'training', eg. in the example used by Mr. French, if your parser
identified
> "Essence" as an attribute, it merely adds it to the table and goes on.
>
> Obviously during training some silly configurations will appear. Many can
be
> caught by just examining the various tables.
>
> -ralph
>
>

"Ivan Debono" <ivanm***@hotmail.com> wrote in message
news:OCi$%238zyFHA.2556@TK2MSFTNGP10.phx.gbl...
>
> "Ralph" <nt_consultin***@yahoo.com> schrieb im Newsbeitrag
> news:3JidnUb_G7Lw7dveRVn-hg@arkansas.net...
> >
> > "Ralph" <nt_consultin***@yahoo.com> wrote in message
> > news:orydnShpNqGM8NvenZ2dnUVZ_smdnZ2d@arkansas.net...
> > >
> > > "Ivan Debono" <ivanm***@hotmail.com> wrote in message
> > > news:eB9hUzyyFHA.736@tk2msftngp13.phx.gbl...
> > > > Ahh, accidentally posted without continuing!!
> > > >
> > > > Here's the full post again:
> > > >
> > > > I'm building  a vb6 application with ado 2.81 and access 2003 as a
> > > backend.
> > > > The app parse recipe files and import ingredients into a table.
> > > >
> > > > Now, let's take as an example 8 different recipes that have the same
> > > > ingredient (eg. cilantro). When I parse these 8 recipes I will check
> if
> > > the
> > > > ingredient is not already available in the table. If it is, i'll use
> it
> > in
> > > > the recipe. If not, I have to create a new one. So let's take these
8
> > > > recipes where they have the following ingredient line for cilantro:
> > > >
> > > > Snipped fresh cilantro
> > > > Cilantro -- finely chopped
> > > > Leaves of Cilantro -- optional
> > > > Chopped cilantro
> > > > Cilantro, fresh, chopped
> > > > Cilantro, chopped
> > > > fresh cilantro leaves -- coarsely chopped
> > > > Cilantro, minced
> > > >
> > > > The whole idea of the app is to have only one ingredient called
> cilantro
> > > and
> > > > not 8. How would one go about designing an intelligent function that
> > > > recognizes that cilantro is the ingredient and will use that instead
> of
> > > > creating 8 different records in the table?
> > > >
> > > > Thanks,
> > > > Ivan
> > > >
> > >
> > > Quick First Brush White Board
> > >
> > > // table_Recipe   DDL
> > > ID:            seq_auto_number
> > > Name:       Text
> > > ' DataView
> > > 1    Soup
> > > 2    Meatloaf
> > >
> > > // table_RecipeIngredientsXREF
> > > RecipeID:      FK (number)
> > > IngredID:        FK (number)
> > > ' dataview
> > > 1    2
> > > 1    3
> > >
> > > // table_Ingredient  DDL
> > > ID:             seq_auto_number
> > > Item:          FK (number)
> > > Source        FK (number)
> > > Preparation: FK (number)
> > > ' dataview
> > > 1    1     1    1
> > > 2    2     2    5
> > > 3    3     1    1
> > >
> > > // table_Item    DDL
> > > ID:        seq_auto_number
> > > Name:    Text
> > > ' DataView
> > > 1 Cilantro
> > > 2 Onion
> > > 3 Celery
> > >
> > > // table_Source   DDL
> > > ID:            seq_auto_number
> > > Source:       Text
> > > ' DataView
> > > 1    Fresh
> > > 2    Canned
> > > 3    Bagged
> > >
> > > // table_Preparation   DDL
> > > ID:            seq_auto_number
> > > Prepare:    Text
> > > ' DataView
> > > 1    chopped
> > > 2    corsley chopped
> > > 3    snipped
> > > 4    whole
> > > 5    minced
> > >
> > > Should give you an idea.
> > > Note: You may not want to use an auto_number
> > > A complete unique readable tag would do.
> > > There are pros and cons.
> > >
> > > hth
> > > -ralph
> > >
> >
> > Forgot to add that the tables would serve as both storage for "lookup"
> keys
> > and also for the "gathering". A true expert system will also include
> > 'training', eg. in the example used by Mr. French, if your parser
> identified
> > "Essence" as an attribute, it merely adds it to the table and goes on.
> >
> > Obviously during training some silly configurations will appear. Many
can
> be
> > caught by just examining the various tables.
> >
> > -ralph
> >
> >
>
> I've got all the tables... I just need to train it :)
>
> Ivan
>

"Ivan Debono" <ivanm***@hotmail.com> schrieb im Newsbeitrag
news:OQmj5qLzFHA.2960@tk2msftngp13.phx.gbl...
> >
> > I can see that from your replies to the others. It wasn't obvious to me
as
> > to how far you had gotten in your design, from you OP. Apologize if I
was
> > too pedantic.
> >
> > -ralph
> >
> >
>
> No problem.
>
> I also included to features after posting.
>
> 1. A Soundex function. Compares the soundex codes and assigns a rating
> 2. A rating system by breaking down the sentence into words and comparing
> those.
>
> I sort by rating in desc order will give the highest rating. 98% of the
time
> it's really accurate but if fails miserably in the remaining 2%.
>
> You guys would do something similar or an alternative rating system?
>
> Ivan
>
>

"Ivan Debono" <ivanm***@hotmail.com> wrote in message
news:eoilgwLzFHA.2008@TK2MSFTNGP10.phx.gbl...
>
> "Ivan Debono" <ivanm***@hotmail.com> schrieb im Newsbeitrag
> news:OQmj5qLzFHA.2960@tk2msftngp13.phx.gbl...
> > >
> > > I can see that from your replies to the others. It wasn't obvious to
me
> as
> > > to how far you had gotten in your design, from you OP. Apologize if I
> was
> > > too pedantic.
> > >
> > > -ralph
> > >
> > >
> >
> > No problem.
> >
> > I also included to features after posting.
> >
> > 1. A Soundex function. Compares the soundex codes and assigns a rating
> > 2. A rating system by breaking down the sentence into words and
comparing
> > those.
> >
> > I sort by rating in desc order will give the highest rating. 98% of the
> time
> > it's really accurate but if fails miserably in the remaining 2%.
> >
> > You guys would do something similar or an alternative rating system?
> >
> > Ivan
> >
> >
>
> Would it make sense to implement regular expressions and what would the
> search string look like?
>
> Ivan
>

"Ralph" <nt_consultin***@yahoo.com> schrieb im Newsbeitrag
news:58adnZTIRONJstTeRVn-jg@arkansas.net...
>
> "Ivan Debono" <ivanm***@hotmail.com> wrote in message
> news:eoilgwLzFHA.2008@TK2MSFTNGP10.phx.gbl...
> >
> > "Ivan Debono" <ivanm***@hotmail.com> schrieb im Newsbeitrag
> > news:OQmj5qLzFHA.2960@tk2msftngp13.phx.gbl...
> > > >
> > > > I can see that from your replies to the others. It wasn't obvious to
> me
> > as
> > > > to how far you had gotten in your design, from you OP. Apologize if
I
> > was
> > > > too pedantic.
> > > >
> > > > -ralph
> > > >
> > > >
> > >
> > > No problem.
> > >
> > > I also included to features after posting.
> > >
> > > 1. A Soundex function. Compares the soundex codes and assigns a rating
> > > 2. A rating system by breaking down the sentence into words and
> comparing
> > > those.
> > >
> > > I sort by rating in desc order will give the highest rating. 98% of
the
> > time
> > > it's really accurate but if fails miserably in the remaining 2%.
> > >
> > > You guys would do something similar or an alternative rating system?
> > >
> > > Ivan
> > >
> > >
> >
> > Would it make sense to implement regular expressions and what would the
> > search string look like?
> >
> > Ivan
> >
>
> Regular expressions definitely makes pattern-matching searches easier, but
> it carries a lot of overhead. Though it simplifies the 'code' - you write
a
> pattern and submit something to be 'parsed' - there is a very complex
engine
> running underneath. The advantage is you didn't have to write the engine.
So
> they are useful in that regard. I would definitely use them in the
> beginning. Saves a lot of writing, but it is only an implementation tool.
>
> Note RE only works with 'patterns', ie, you have to have a discernable
> pattern before they do you any good. Obviously a recipe that contained a
> bulleted list of ingredients at the top or other defined structure would
> lend itself to defining a pattern, but what about some plain text that add
s
> ingredients as it rattles on? OSFA will not apply.
>
> Your task can be broken down into Input massage -> Knowledgebase -><-
> Forward Chaining -><- Inference Engine -><-- Backward Chaining -> Output.
> You will likey use a ton of different tools as you work your way through
the
> process. There is a lot of useful information on the net and several books
> on the subject. It can get complicated.
>
> Don't leave out the chaining. Automate the 'learning' whenever possible -
> else you will find yourself merely creating a linear procedural process
with
> a ton of conditionals. <g>
>
> But take heart that what you are doing is creating in a sense an 'expert'
> system. No two solutions are ever the same - depends on what 'expertise'
is
> needed in the output. Also by the time you are done, you will likely be
> eligible for a good job at Google.
>
> My only exposure to AI/Expert systems is in training and backward chaining
> (Neural Nets). The knowledgebase and Inference engines were rather well
> defined before I got there. (Thank ghod! <g>) When you get there give me a
> call. <smile>
>
> Enjoy. I can't think of a more fascinating project. I know I thoughly
> enjoyed every one that I have been involved in and missed it when I left.
>
> -ralph
>
>
>

"Ivan Debono" <ivanm***@hotmail.com> wrote in message
news:%23wvXoXWzFHA.3256@TK2MSFTNGP09.phx.gbl...
>
> "Ralph" <nt_consultin***@yahoo.com> schrieb im Newsbeitrag
> news:58adnZTIRONJstTeRVn-jg@arkansas.net...
> >
> > "Ivan Debono" <ivanm***@hotmail.com> wrote in message
> > news:eoilgwLzFHA.2008@TK2MSFTNGP10.phx.gbl...
> > >
> > > "Ivan Debono" <ivanm***@hotmail.com> schrieb im Newsbeitrag
> > > news:OQmj5qLzFHA.2960@tk2msftngp13.phx.gbl...
> > > > >
> > > > > I can see that from your replies to the others. It wasn't obvious
to
> > me
> > > as
> > > > > to how far you had gotten in your design, from you OP. Apologize
if
> I
> > > was
> > > > > too pedantic.
> > > > >
> > > > > -ralph
> > > > >
> > > > >
> > > >
> > > > No problem.
> > > >
> > > > I also included to features after posting.
> > > >
> > > > 1. A Soundex function. Compares the soundex codes and assigns a
rating
> > > > 2. A rating system by breaking down the sentence into words and
> > comparing
> > > > those.
> > > >
> > > > I sort by rating in desc order will give the highest rating. 98% of
> the
> > > time
> > > > it's really accurate but if fails miserably in the remaining 2%.
> > > >
> > > > You guys would do something similar or an alternative rating system?
> > > >
> > > > Ivan
> > > >
> > > >
> > >
> > > Would it make sense to implement regular expressions and what would
the
> > > search string look like?
> > >
> > > Ivan
> > >
> >
> > Regular expressions definitely makes pattern-matching searches easier,
but
> > it carries a lot of overhead. Though it simplifies the 'code' - you
write
> a
> > pattern and submit something to be 'parsed' - there is a very complex
> engine
> > running underneath. The advantage is you didn't have to write the
engine.
> So
> > they are useful in that regard. I would definitely use them in the
> > beginning. Saves a lot of writing, but it is only an implementation
tool.
> >
> > Note RE only works with 'patterns', ie, you have to have a discernable
> > pattern before they do you any good. Obviously a recipe that contained a
> > bulleted list of ingredients at the top or other defined structure would
> > lend itself to defining a pattern, but what about some plain text that
add
> s
> > ingredients as it rattles on? OSFA will not apply.
> >
> > Your task can be broken down into Input massage -> Knowledgebase -><-
> > Forward Chaining -><- Inference Engine -><-- Backward Chaining ->
Output.
> > You will likey use a ton of different tools as you work your way through
> the
> > process. There is a lot of useful information on the net and several
books
> > on the subject. It can get complicated.
> >
> > Don't leave out the chaining. Automate the 'learning' whenever
possible -
> > else you will find yourself merely creating a linear procedural process
> with
> > a ton of conditionals. <g>
> >
> > But take heart that what you are doing is creating in a sense an
'expert'
> > system. No two solutions are ever the same - depends on what 'expertise'
> is
> > needed in the output. Also by the time you are done, you will likely be
> > eligible for a good job at Google.
> >
> > My only exposure to AI/Expert systems is in training and backward
chaining
> > (Neural Nets). The knowledgebase and Inference engines were rather well
> > defined before I got there. (Thank ghod! <g>) When you get there give me
a
> > call. <smile>
> >
> > Enjoy. I can't think of a more fascinating project. I know I thoughly
> > enjoyed every one that I have been involved in and missed it when I
left.
> >
> > -ralph
> >
> >
> >
>
> Did you develop the systems in vb?
>
> Ivan
>