Information Retrieval Facility

from Wikipedia, the free encyclopedia
IRF logo

The Information Retrieval Facility (short IRF ) was a research platform and served the collaboration of experts in the field of Information Retrieval (IR).

It was founded in 2006 and was based in Vienna . The IRF was the world's first e-science system that was exclusively dedicated to the semantic processing of text. The staff included experts, researchers and students in the fields of information retrieval and information management .

The IRF ceased its activities in 2012.

aims

The Information Retrieval Facility was used for research and creation of scientific goals. This included the modeling of information retrieval systems for global patent document collections. Complex searches can be carried out with these systems, which can be translated as information recovery. In addition to text files, this also includes information from images .

Another branch of the company was the research and development of a technical infrastructure that enables interactive experiments with formal and mathematical retrieval concepts for very large document collections. Closely related to this is the investigation of the usability of multimodal user interfaces of large-scale information retrieval systems in order to enable appropriate operation by different user groups. Furthermore, the integration of users and their needs was included in the process of modeling information retrieval systems so that an accurate performance evaluation was ensured.

Patent data were shown in different views so that the focus is guaranteed in the corresponding dependency.

The IRF has also advocated the definition of standardized methods for evaluating the information retrieval processes in the patent collections. They wanted to be able to handle text and non-text portions of a patent in a coherent manner and develop search engines that would make it possible to find structured and semi-structured documents in very large patent collections. Assessments should be made as part of the trial. The temporal dimensions of patent documents should be integrated into retrieval strategies.

Another goal of the IRF was to improve the efficiency and precision of patent retrieval based on ontologies and different language techniques and to create improved IR methods that enable the use of unstructured queries within a patent document. Formal (mathematical) identification and specification of business-relevant information is to help intellectual property to detect (Intellectual Property). In addition, the research into scaling mechanisms in the information retrieval area, taking into account the characteristics of patent data and the determination and experimentation with computing architectures for very high capacity information management, was advanced.

The creation of an open e-science platform, which enables the creation and implementation of IR experiments on a common research infrastructure in a uniform and simple manner, was still on the company's agenda. In addition, there was the discovery and research of application purposes and business applications that result from intellectual property information. For this purpose, formal information retrieval, languages ​​and semantic processing in the field of applied sciences, which bring the information into a global, industrial context, were activated . The development and integration of various information access methods and research into effective methods for interactive information retrieval was another field of activity.

Semantic supercomputing

Current technologies for extracting concepts from unstructured documents are associated with intensive computing power. In order to allow interactive experimentation with large text corpora , the IRF had a high-performance computing (HPC) environment for high- performance text mining . This was equipped with a multi-node system consisting of 80 cores that could be increased to 1024 cores. This was associated with a high-speed interconnect technology. There were also individual systems with large storage options of 320 GB, which could be expanded to 4 TB. Thanks to 4 FPGA cores, which could be expanded up to 256 cores, the systems were also completely switchable.

World Patent Corpus

The aim of the IRF was to create a platform for patent experts based on modern information retrieval technologies. It was expected that information retrieval (IR) technologies would take center stage in information technology.

The entirety of all patent documents represents a huge body of text. Patents have become a crucial topic, especially for global companies and universities. The industrial users of patent data are among the most demanding and important information professionals of all. These audiences will benefit most from technology that helps them explore large amounts of data.

Web links

credentials