Bored of text boxes where you ask questions in natural language, on systems that respond slowly, or incorrectly? Dylan-Q is an intuitive guiding system for building questions.
All the questions that can be built with the system provide a correct answer, and questions that can not be constructed (a minimum percentage of the questions that users ask) are treated in a conventional way or, if considered relevant, can be included in the system.
The system has been tested with real data from scientific conferences of the publisher Springer and with heavily populated domains of DBpedia, demonstrating its flexibility when dealing with complex data models and with very large datasets. The tests performed show users very satisfied with the ease of constructing the questions, the response times and the answers retrieved.
It allows customers to formulate questions in natural language intuitively and quickly. The questions are grammatically correct, flexible (allows asking in different ways), and multilingual (English, Spanish or German).
The results of the question are updated as the question is being constructed and refined. This gives the user a great sense of control and confidence.
This tool is the result of a combination of the latest scientific advances and an advanced R&D+i, in human-computer interaction, semantics, and computational linguistics, by researchers from UPM and Bielefeld Univ. (Germany).
The technologies developed are in the process of creating some "objects of protection" (patents, sw registration and industrial secret). Demonstrations can be made under a confidentiality agreement.
“Intuitive, multilingual and personalized solution to create questions using the terminology and data of each company”
"This tool allows you to query data repositories in a simple, intuitive and efficient way"
- "53% of US adults quit an online purchase if they do not find a quick answer to their questions."
- "73% say that ‘valuing your time’ is the best thing a company can do to give a good service online”.
-Improvement of 1.9% of European Gross Domestic Product due to Big & Open data.
-An increase of 5-6% in the effectiveness of decisions based on data.
Mariano Rico, PhD.
Innovation, Commercialization and Entrepreneurship Area
Centre of Support for Technological Innovation – UPM