Known issues and limitations

The following are the limitations of the IBM Content Assistant:

• The IBM Content Assistant cannot extract text from images, whether typed or handwritten. For such document content, you need to first use IBM Datacap or Business Partner solutions to extract text from such documents. You can then use the Content Assistant with the extracted documents to generate document summaries or answer questions.
• The IBM Content Assistant poses a limitation when it extracts data from PDF forms like ArcoForms or XFA Forms. It is not able to associate the values of form fields with the appropriate name fields. This limitation impacts the usefulness of document summarization for the form fields and the assistant does not provide desired answers for questions related to such form-related document content.
• When you ask questions to the Content Assistant about PDF documents, the response includes relevant excerpts from the source document. You can click on an excerpt to open the corresponding source document. However, when you close the source document and exit the excerpt view within Navigator, the Navigator interface renders a blank page.