Skip to Main Content
IBM Data and AI Ideas Portal for Customers

Shape the future of IBM!

We invite you to shape the future of IBM, including product roadmaps, by submitting ideas that matter to you the most. Here's how it works:

Post your ideas

Post ideas and requests to enhance a product or service. Take a look at ideas others have posted and upvote them if they matter to you,

  1. Post an idea

  2. Upvote ideas that matter most to you

  3. Get feedback from the IBM team to refine your idea

Help IBM prioritize your ideas and requests

The IBM team may need your help to refine the ideas so they may ask for more information or feedback. The product management team will then decide if they can begin working on your idea. If they can start during the next development cycle, they will put the idea on the priority list. Each team at IBM works on a different schedule, where some ideas can be implemented right away, others may be placed on a different schedule.

Receive notification on the decision

Some ideas can be implemented at IBM, while others may not fit within the development plans for the product. In either case, the team will let you know as soon as possible. In some cases, we may be able to find alternatives for ideas which cannot be implemented in a reasonable time.

Additional Information

To view our roadmaps:

Reminder: This is not the place to submit defects or support needs, please use normal support channel for these cases

IBM Employees:

The correct URL for entering your ideas is:



Natural Language Understanding (NLU)

Showing 28

Additional language support

We really need the ability to use NLU (classification, sentiment, keyword extraction or entities) in additional languages including Vietnamese and Thai, Traditional Chinese (HK Cantonese)
5 months ago in Natural Language Understanding (NLU) 0 Submitted

NLU PT-BR model

The NLU Service has wrong (and very) misinterpreted feelings for some very trivial phrases, following some recent examples of phrases that it returns "0". • Pelo bom atendimento da Loja e pela qualidade dos carros. • Carros muito confiáveis, si...
over 1 year ago in Natural Language Understanding (NLU) 0 Submitted

Source IP address filtering feature for "public" Watson services

Our company requires every external applications service to add an additional authentication method other than the apikey authentication. For example: token based 2 factor authentication, client certificate authentication or source IP address filt...
over 2 years ago in Natural Language Understanding (NLU) 0 Future consideration

anti-synonyms for entities in Watson

We have synonyms in watson, and fuzzy matching - which works great! But there are cases were Watson selects completely wrong entities. I have no way of fixing this. My example case:- user text contains the string 'Whats' - watson identifies...
over 2 years ago in Natural Language Understanding (NLU) 1 Functionality already exists

Search fields from specific datasets

Regarding Cognos Analytics functionality NLP - Natural Language Process. While searching fields using NLP, we get results from entire Cognos server which takes lot of time to search. Can we limit the search option to the Data source/Data Set that ...
over 2 years ago in Natural Language Understanding (NLU) 1 Not under consideration

Make emotion analysis with NLU available in German

Developing further the semantic analysis of German text input should be considered for enriching the emotion feature and differentiating from other NLP tools. Currently the emotion feature of IBM Watson NLU is only supporting English. However, mo...
over 2 years ago in Natural Language Understanding (NLU) 0 Submitted

Revert NLU model to return neutral sentiment scores

Whereas we previously saw that the NLU sentiment scores were distributed throughout the -1 to +1 scale, in the last several months I’ve seen that there are virtually no neutral scores assigned. In the past it was very common for NLU to assign scor...
over 2 years ago in Natural Language Understanding (NLU) 1 Not under consideration

Ability to return sub section of identified regex as entity

We can identify any regular expression but can we *return specific part of extraction as entity* .e.g. regex is TN(\s)?(#)?(\s)?[0-9]{6,10} , I wish to identify whole expression and return only last identified number. TN # 12345689 From about ex...
about 3 years ago in Natural Language Understanding (NLU) 0 Submitted

Faster response from NLU RBR every time.

RBR models tend to give Downstream error or loads late for the first time after delay like 15 min. Custer feels RBR is unstable and inconsistent which is reducing the use of RBR model. Where as ML model does not show any such issues, Ideally we...
about 3 years ago in Natural Language Understanding (NLU) 0 Submitted

Deploy SDU as a Service through a NLU enrichment

SDU brings many use case possibilities, however being inside WDS, is limited to use cases where documents are stored. With the ability to deploy the SDU model to NLU, the outputs can be used in a more flexible way.
over 3 years ago in Natural Language Understanding (NLU) 0 Future consideration