We performed a comparison between HyperScience and UiPath Document Understanding based on real PeerSpot user reviews.
Find out in this report how the two Intelligent Document Processing (IDP) solutions compare in terms of features, pricing, service and support, easy of deployment, and ROI."One of the most valuable features of HyperScience is the user-training module. Whenever the extraction takes place, based on the way we have trained HyperScience, it would give us some success status or a certain confidence level. If the solution has processed something that it determined was not extracted correctly it will queue those items for manual review."
"I like that compared to other tools, HyperScience works best with handwritten documents."
"Has algorithms that can detect a document template even if the image has a lot of distortions."
"We have seen pretty good accuracy."
"It provides the best accuracy for handwritten forms, which is a struggle in the industry. You can take processes with a lot of manual work and streamline them through this tool."
"Valuable features include tools like IQ Bot and the ability to extract handwritten documents with 93-95 per cent accuracy."
"What I liked more about HyperScience was the quality of the OCR it is a lot better compared to Google."
"The most valuable feature is key-value pair and table extraction."
"With Document Understanding, tasks that would normally take eight hours manually can be completed in three to five hours. It still requires some human intervention, but 90 percent of the processing can be automated."
"It's great for document understanding for invoices and installments."
"UiPath Document Understanding is user-friendly, with an easy-to-use self-trained model, and the OCR it provides does a good job even with scanned PDFs."
"The most valuable feature in UiPath Document Understanding is the identification of the fields column in the PDF documents."
"OCR technology is undoubtedly the most valuable feature and the feasibility of integrating data processes with AI and machine learning models is fascinating."
"It's helped us free up time for other staff projects."
"The taxonomy and Validation Station are among the most helpful features for us. If anything is extracted incorrectly, we can manually extract it there."
"Extracting tables from certain documents could be improved."
"HyperScience could improve the unstructured data extraction feature."
"No solution is perfect and there are several different scenarios that could be improved in HyperScience. One area is where there are multiple tables in the same form I have seen HyperScience struggle. There is some issue with supporting the extraction from multiple tables involved on the same form. If this could improve, it would be a big benefit."
"The product's usability could be better. The first pain point is that we're getting the output in a different format, and we were expecting a different timetable. The second point is that if you want better results, HyperScience says you have to configure a minimal PDF or a maximum of 400 PDFs. If you want results with 400 PDFs for what's written by these doctors, then you also configure the maximum of 400 templates for that. So, it's essentially a lack of support from HyperScience. In the next release, it would be better if failure scenarios were reduced. It would also help if they offered different formats, inputs or injections, and added different scenarios."
"The solution lacks support for a greater range of languages."
"HyperScience has less capability while working on unstructured forms. Unstructured forms are those where there is no standard structure and the information can be anywhere on the form. They need to develop this capability."
"They could work on the price and make it a bit more reasonable."
"Currently, we have to train multiple templates because the column size and row size change in each PDF."
"The licensing model poses a significant challenge due to the fee charged for posting a model, which impedes the development of productivity-enhancing models."
"There is room for improvement in UiPath Document Understanding's pricing. It is expensive for small clients. Currently, there is a big gap between the basic package and the 200,000 packages. There is no package in the middle for small agencies."
"The coding machine learning could be a little bit better."
"The documentation should be more clear, or better training should be provided."
"Its pricing can be improved."
"The solution must localize the built-in features for supporting Arabic scripts so we do not rely on third-party tools."
"It would be much easier if UiPath increased the count of pages. Currently, they are allowing one million pages for $10,000 per month. I would prefer to increase the page count or reduce the dollar count in terms of processing the documents. I would prefer $6,000 per month for processing 2 to 3 million pages per month. It will then be much easier for companies with a low budget to use this product."
More UiPath Document Understanding Pricing and Cost Advice →
HyperScience is ranked 5th in Intelligent Document Processing (IDP) with 7 reviews while UiPath Document Understanding is ranked 3rd in Intelligent Document Processing (IDP) with 45 reviews. HyperScience is rated 7.6, while UiPath Document Understanding is rated 8.2. The top reviewer of HyperScience writes "It has a lot of functionality, whatever we use, but a few things could be improved". On the other hand, the top reviewer of UiPath Document Understanding writes "Is easy to configure, user-friendly, and produces accurate results". HyperScience is most compared with ABBYY Vantage, UiPath, Instabase, Microsoft Power Automate and Automation Anywhere (AA), whereas UiPath Document Understanding is most compared with ABBYY Vantage, Instabase, Tungsten TotalAgility, Nanonets and Datamatics TruCap+. See our HyperScience vs. UiPath Document Understanding report.
See our list of best Intelligent Document Processing (IDP) vendors.
We monitor all Intelligent Document Processing (IDP) reviews to prevent fraudulent reviews and keep review quality high. We do not post reviews by company employees or direct competitors. We validate each review for authenticity via cross-reference with LinkedIn, and personal follow-up with the reviewer when necessary.
