We have a request to create spend categories based on the analysis of free text data. This may be a combination of Omni functions i.e. text analysis and machine learning.
Are there any guidelines or can anyone provide some pointers as to the best way to manage this?
Regards
Karen
0 Votes
7 Comments
Sorted by
K
Karen van Stadenposted
almost 2 years ago
Hi Paola
I posted a data sample for text analysis. Are you able to advise of the most efficient way to manage this?
Regards
Karen
0 Votes
S
Susanposted
almost 2 years ago
Not a problem, there are a lot out there I'd just be careful about what they can promise you, hopefully they can give you a sample..
Good luck :)
0 Votes
K
Karen van Stadenposted
almost 2 years ago
Hi Susan
Thank you for taking the time to respond.
We are looking to automate the process as far as possible as there will potentially be high data volumes. We are also bidding against an organisation that provides automatic classification a standard.
Many thanks
Karen
0 Votes
S
Susanposted
almost 2 years ago
Hi Karen,
This is pretty much what we do for a living, it looks like it might be credit card or expense data. We tend to classify this manually as there's too much that can go wrong with keyword searches with things like this. Here's a couple of videos I've made showing what we do:
Hi Karen, we're The Classification Guru and use Omniscope for this very thing! We use it in a slightly different way for spend classification but always recommend starting by doing it manually the first time with people before automating the process. If you want to have a chat, you can email me susan@theclassificationguru.com.
0 Votes
P
Paola Tomeiposted
almost 2 years ago
Admin
Hello,
Omniscope is used by many spend analysis specialists for this purpose...
Are you able to post a small sample of your dataset (you can scramble/remove anything sensitive) and just show what is the text to be used for this classification?
We have a request to create spend categories based on the analysis of free text data. This may be a combination of Omni functions i.e. text analysis and machine learning.
Are there any guidelines or can anyone provide some pointers as to the best way to manage this?
Regards
Karen
0 Votes
7 Comments
Karen van Staden posted almost 2 years ago
Hi Paola
I posted a data sample for text analysis. Are you able to advise of the most efficient way to manage this?
Regards
Karen
0 Votes
Susan posted almost 2 years ago
Not a problem, there are a lot out there I'd just be careful about what they can promise you, hopefully they can give you a sample..
Good luck :)
0 Votes
Karen van Staden posted almost 2 years ago
Hi Susan
Thank you for taking the time to respond.
We are looking to automate the process as far as possible as there will potentially be high data volumes. We are also bidding against an organisation that provides automatic classification a standard.
Many thanks
Karen
0 Votes
Susan posted almost 2 years ago
Hi Karen,
This is pretty much what we do for a living, it looks like it might be credit card or expense data. We tend to classify this manually as there's too much that can go wrong with keyword searches with things like this. Here's a couple of videos I've made showing what we do:
https://youtu.be/GqwxZf89Wy4
https://youtu.be/VSMwTYVPMes
0 Votes
Karen van Staden posted almost 2 years ago
Hi
Attached is a sample dataset. The classification is to be based on the 'linedescription'
Regards
Karen
Attachments (1)
FreeText Cla....xlsx
1.67 MB
0 Votes
Susan posted almost 2 years ago
Hi Karen, we're The Classification Guru and use Omniscope for this very thing! We use it in a slightly different way for spend classification but always recommend starting by doing it manually the first time with people before automating the process. If you want to have a chat, you can email me susan@theclassificationguru.com.
0 Votes
Paola Tomei posted almost 2 years ago Admin
Hello,
Omniscope is used by many spend analysis specialists for this purpose...
Are you able to post a small sample of your dataset (you can scramble/remove anything sensitive) and just show what is the text to be used for this classification?
Thanks!
0 Votes
Login or Sign up to post a comment