ORDO – Order in a digital world

Human knowledge is available in many forms, with information stored in a variety of data media and formats as texts, images or video and audio recordings. Researchers involved in the ORDO application scenario are developing a system that will make it easier to manage this diversity.

Modern information technology makes our lives easier, but sometimes more confusing as well. This can be seen, for example, in the development of new medications. Today, pharmacologists seeking information about active ingredients find that the search can be a complicated and drawn-out process. They may find chemical formulas in specialized databases and information on how medications work in pharmaceutical archives, while finding a complete list of medications with illustrations may require consulting a manufacturer’s marketing materials.

A better way to organize knowledge
To make searching for information easier and more efficient, ORDO researchers are developing new services and software tools to help users organize and prioritize contents and search results. The idea is to provide users with all of the relevant data and help them distinguish quickly between important and unimportant information and recognize how one thing relates to another. New technologies will allow pharmaceutical researchers to link searches in different databases and sources with a central query, and to coordinate the search process. For example, the computer might search a database for a chemical formula and find the corresponding molecular diagram in a different source. When it discovers a mention of an active ingredient on a drug package, it automatically deduces that the medication is important in that context – and recognizes why it is being used. All of this information is clearly presented to researchers so that they can organize the search results appropriately and select certain items about which they want to learn more.

This overview of the available information is automatically generated. Procedures are also being developed to process large quantities of data and display them in graphic form. The result is comprehensive, individualized knowledge management that consistently allows the user to find the right information.

Practical testing in the context of research and development
ORDO’s research results are being tested specifically to determine whether they meet the needs of research and development departments, such as those in the chemical and pharmaceutical industries, which need to handle large amounts of data. In addition to primary data, such as patents, a great deal of secondary information – research reports, market analyses and other academic publications – needs to be analyzed as well. E-mails, archives, in-house wikis and blogs also have to be processed. ORDO’s researchers are developing methods and technologies that will make it possible to use all of these data resources more efficiently and link them with Internet content.

Managing information using semantic technologies
With ORDO, computers will soon be able to identify concepts and relationships in text documents, extract facts and generate summaries automatically. Relative to existing recognition methods, this represents a major step toward more efficient information management. In the future, these technologies will also be used for analyzing and organizing multimedia content.

This ambitious project is based on the enhancement of semantic technologies that enable a computer to automatically recognize data content and meaning, and to put data into context. This is particularly important when organizing and analyzing unstructured data. Another focus of ORDO is to develop efficient procedures for creating classification systems (such as ontologies) for modeling, managing and organizing data. With the help of innovative linguistic and statistical methods, these ontologies can be used still more effectively, and in some cases they may even be generated automatically.


The key is personalization
To achieve truly individualized knowledge management, it is important to take into account personal information and perspectives – for example, specific work areas and locally used computers. The kinds of information that are relevant to a project manager, for example, will differ from the information a member of the project staff might need. Preparing and organizing data in a personalized way will make it possible to give due consideration to the individual needs and preferences of each user.