Frns,

Any more tries / suggestions?

On Tue, Jun 7, 2011 at 11:26 PM, vishnu madiga <[email protected]>wrote:

> Good suggestion Navneet..  I had a similar solution but am looking for
> something more faster.
> Performance is a major consideration to me. The possible permutations for a
> large set of debits and credits could be massive. The solution should be
> intelligent enough to discard non-possible branches quickly.
>
>
>
>
> On Tue, Jun 7, 2011 at 11:15 PM, Navneet Gupta <[email protected]>wrote:
>
>> Well, one approach is have all possible sums stored for debits and
>> credits and match the sums.
>>
>> Basically, you can create a multi-list type DS (if i remember the name
>> correctly) and store the SUMS and with formative elements in linked
>> lists being pointer by node containing sums
>>
>> SUM 1 -> Header 1 -> Header 2 -> Header 3
>>
>> Header1 -> Debit m -> Debit n (Here one of the possibilities of
>> forming a sum SUM 1 is Debit m + Debit n)
>>
>> Now, it's basically a design choice whether you want to have a single
>> DS like above to contain both Credit and Debit based calculations or
>> have separate multi-lists for Debits and Credits.
>>
>>
>> On Tue, Jun 7, 2011 at 7:39 PM, vishnu madiga <[email protected]>
>> wrote:
>> > find a combination of credits and debits which will sum out to zero in
>> > a given set of credit and debit values.
>> > It's not necessary to have a one to one match between a debit and
>> > credit.
>> >
>> > For example, the parent may lend $25 and $75 to its subsidiary and
>> > receive back 3 payments of $20, $40 and $40.
>> > The task is to match these entries in an account. This becomes taxing
>> > because multiple debits can be applied to
>> > multiple credits. In many cases there is more than one way to match
>> > the values.
>> >
>> > The goal is to develop a solution which will take a set of credits and
>> > debits and match as many debits to as
>> > many credits as possible. A test case could have up to 2,000 debits
>> > and credits.
>> >
>> >
>> > --
>> > You received this message because you are subscribed to the Google
>> Groups "Algorithm Geeks" group.
>> > To post to this group, send email to [email protected].
>> > To unsubscribe from this group, send email to
>> [email protected].
>> > For more options, visit this group at
>> http://groups.google.com/group/algogeeks?hl=en.
>> >
>> >
>>
>>
>>
>> --
>> --Navneet
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Algorithm Geeks" group.
>> To post to this group, send email to [email protected].
>> To unsubscribe from this group, send email to
>> [email protected].
>> For more options, visit this group at
>> http://groups.google.com/group/algogeeks?hl=en.
>>
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Algorithm Geeks" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/algogeeks?hl=en.

Reply via email to