InfoTrie Labs is a tool allowing to exploit and to explore the documentary bases of the companies. It uses Deep Learning techniques to OCR and extract structured information from documents to enable intelligent, natural language searching across your entire database.

Challenges

Companies all have dormant mines of information in their documentary databases. Whether it is information contained in technical documentation, PDF documents or results of experiments stored for years, this information can prove invaluable for the performance of the company. Searching for information in a set of documents can therefore be costly in terms of time and resources for the company.

Goals

InfoTrie Labs is a solution for analyzing your document base and querying it in natural language. Thus, our tool allows you to analyze all types of documents (standardized or non-standardized), to extract the relevant information and to store it in a structured way in a database.

Plus

  • A powerful tool for analyzing documents in different formats (pdf, png, excel, word, pptx) with identification of key areas (figures, tables, text, etc.),
  • Extraction of information contained in these key areas (tables, text, figures),
  • An intelligent tagging tool to identify textual objects (proper names, organizations, location, etc.) adaptable to a specific business vocabulary, keywords, corpus structure (title, section, paragraph),
  • Storage of extracted information (tags, text, figures, tables, etc.) in a structured storage space,