Always find the information you are looking for

 

Accurate access to documents containing answers to your questions

Automatyczne Odczytywanie treści dokumentu (OCR)

Automatyczna interpretacja tekstu

Grupowanie / klasyfikacja dokumentów

Wyszukiwanie powiązań

  • PRODUCTS
  • DoRIS

DoRIS text analysis solution facilitating access to unlimited resources of knowledge accumulated in the form of unstructured documents, stored in local resources and in private and public clouds.

Minimize the time to reach information.
Centralize access to distributed information resources.
Reduce the labor intensity of document production.
Efficient navigation of knowledge resources.
DoRIS

Document Repository Intelligent Search

A system that through artificial intelligence and machine learning mechanisms:

Understands the contents of documents stored in its database. Document content is automatically analyzed for occurrences of named entities (naming units), identification data, numerical data in tables, image content. Key phrases and abstracts are determined. 

It learns and adapts its operation to the organization's processes based on user interaction.  It can supplement automatically generated document information, for example, by assigning tags.

It provides quick access to information through powerful indexes and contextual information retrieval mechanisms that supplement traditional full-text search mechanisms with context derived from the semantics of the search phrases.

DoRIS allows organizations to move from manual, time-consuming work to reading text automatically and making quick decisions. It will point out places in the documents that are important to the decision maker. 

Need other solutions related to text analytics?

DoRIS Functionalities

The development of the algorithms used in the system is supported by a research and development project co-financed by the NCBiR

Intelligent OCR

Recognizes text from the page image taking into account embedded elements such as tables, images, headers and footers present in the document. The system automatically builds the content of the document, correcting errors resulting from the low resolution of the document.

Metadata

Extract metadata such as file size, creation and last modification date, author, file format. Metadata can be used to narrow down search results.

Content extraction

Intelligent extraction of document content, taking into account elements of document formatting (headers, footers, embedded objects). Extraction of content from file formats where such extraction is difficult or generates a lot of errors - the system allows you to retrieve it as a text file.

Automatic assignment of key phrases

Traditional approaches based on user-entered keywords do not work because of the tendency of users to minimize effort at the stage of entering information into the database. Incomplete tagged documents are then omitted from search results. The DORIS system automatically generates key phrases thus filling in the gap related to the document description. At each stage of the document's life, the user has the opportunity to verify and complete the key phrases.

Tagging

The user has the ability to tag documents - assigning custom labels that can be used to search for documents.

Automatic cataloging

Automated document cataloging (based on user-defined flexible categories). The system allows you to create your own hierarchical folders, to which the user can then assign documents - this way of navigation reflects the way of working with resources stored locally on the computer, with the user simultaneously organizing his libraries in many ways: by date, subject, document purpose.

Full-text search

Full-text search with word variations, and tolerates errors such as typos. The system supports operators in the body of the query that allow you to enter additional search criteria.

Cognitive search / semantic search

Search for content not only with keywords, but with the meaning of words. The search engine understands the context of the search query. With cognitive search, you can search for content not directly covered by the keywords themselves and described by phrases other than those used in the query, for example: documents containing semantically similar words, concepts, relationships.

Search for similar documents

Search for documents containing content or phrases similar to the model document. A fast search algorithm finds documents that are similar, allowing you to use already existing content. This saves on work that is done only because we are not aware of the existence of results in a particular area.

Eliminate duplicates

Search for documents that exist in the repository in multiple copies (duplicates). The repository often contains a huge number of duplicates, which unnecessarily take up disk space

Document classification

Automatically assign to a case type and start the appropriate document handling process. In companies there are huge amounts of unstructured data (emails, pdf presentations, graphics) that are unused because or require manual search. Automatic classification of documents (into different categories) gives them the right structure, so you do not waste time on manual cataloging /tagging/, and searching in a specific data directory is much easier.

Recommendations

Find before you search. Automatically and expediently suggest to the user other documents that the user may be interested in or that may be useful in performing his/her duties.

Extracting numerical data from tables and forms

The system automatically detects tables on documents and converts their contents to a format that allows their further analysis in spreadsheets.

Automatic summary

Reading a large amount of text and getting to the relevant information takes a lot of time and can be exhausting. With the automatic summary function, only the necessary information is extracted from the text. Maximum content, minimum words - automatic shortening of texts to their most important content/content.

Document repository

By default, the system stores document information, leaving the original files in the source locations. The system can integrate with such file sources as Sharepoint, cloud drives (OneDrive), network folders and others.