An Information Extraction Core System for Real World German Text Processing

An Information Extraction Core System for Real World German Text Processing Gfinter Neumann* Rolf Backofen t Judith Baur t Markus Becker~ Christian Braun ~ Abstract This paper describes SMES, an informa- tion extraction core system for real world German text processing. The basic design criterion of the system is of providing a set of basic powerful, robust, and efficient nat- ural language components and generic lin- guistic knowledge sources which can easily be customized for processing different tasks in a flexible manner. 1 Introduction There is no doubt that the amount of textual infor- mation electronically available today has passed its critical mass leading to the emerging problem that the more electronic text data is available the more difficult it is to find or extract relevant information. In order to overcome this problem new technologies for future information management systems are ex- plored by various researchers. One new line of such research is the investigation and development of in- formation extraction (IE) systems. The goal of IE is to build systems that find and link relevant infor- mation from text data while ignoring extraneous and irrelevant information (Cowie and Lehnert, 1996). Current IE systems are to be quite successfully in automatically processing large text collections with high speed and robustness (see (Sundheim, 1995), (Chinchor et al., 1993), and (Grishman and Sund- heim, 1996)). This is due to the fact that they can provide a partial understanding of specific types of text with a certain degree of partial accuracy using fast and robust shallow processing strategies (basi- cally finite state technology). They have been "made sensitive" to certain key pieces of information and *DFKI GmbH, Stuhlsatzenhausweg 3, 66123 Saarbriicken, Germany, neumannOdfki, uni-sb, de ~LMU, Oettingenstrasse 67, 80538 Miinchen, Ger- many, backofen@informatik, uni-muenchen, de *DFKI GmbH, baur@dfki, uni-sb, de ~DFKI GmbH, mbecker0dfki, uni-sb, de ~DFKI GmbH, cbraun@dfki, uni-sb, de thereby provide an easy means to skip text without deep analysis. The majority of existing information systems are applied to English text. A major drawback of previ- ous systems was their restrictive degree of portabil- ity towards new domains and tasks which was also caused by a restricted degree of re-usability of the knowledge sources. Consequently, the major goals which were identified during the sixth message un- derstanding conference (MUC-6) were. on the one hand, to demonstrate task-independent component technologies of information extraction, and, on the other hand, to encourage work on increasing porta- bility and "deeper understanding" (of. (Grishman and Sundheim, 1996)). In this paper we report on SMES an information extraction core system for real world German text processing. The main research topics we are con- cerned with include easy portability and adaptabil- ity of the core system to extraction tasks of differ- ent complexity and domains. In this paper we will concentrate on the technical and implementational aspects of the IE core technology used for achieving the desired portability. We will only briefly describe some of the current applications built on top of this core machinery (see section 7). 2 …

Content truncated to 500 words