Information Technology Reference
In-Depth Information
4.4.2
CoIN-EF for indiGo
The integration of CoIN-EF into the indiGo platform will be finished in April 2002.
This integration allows to retrieve LLs related to the current process and (project
context) by one click. To do so, the current (project) context is specified and stored
within the user preferences and is stored persistently across the users sessions.
More challenging will be the integration with tools for knowledge construction:
From discourses to experiences and from experiences to process models. As a
preparation, the discourse grammar for the operational phase has been designed
according to the formats for LLs. This should facilitate a mapping of the articles in a
discussion to types of LLs.
4.4.3
Case-Based Reasoning for Sharing Process and Project Knowledge
Since several years there has been a strong tendency in the case-based reasoning
(CBR) community (Kolodner 1993) to develop methods for dealing with more
complex applications. One example is the use of CBR for knowledge management
(KM) (Aha, Becerra-Fernandez et al. 1999). Another one is the integration of CBR
with experience factories (Henninger 1995, Althoff & Wilke 1997, Tautz & Althoff
1997, Bergmann, Breen et al. 1999). The latter also contributed to the development of
the experience management subfield of KM (Tautz 2000, Bergmann 2001, Althoff,
Decker et al. 2001), which already found one implementation through the merger of
the German CBR and KM communities (
www.experience-management.org
,
Minor &
Staab 2002). Meanwhile many papers have been published that are related to the use
of CBR in KM. Weber, Aha, and Becerra-Fernandez (2001) give an overview on
intelligent LLs systems, which includes CBR approaches. While Wargitsch (1998)
describes how CBR can be used for workflow support, Chen-Burger, Robertson, and
Stader (2000) focus on the support for business modeling in general. Decker and
Jedlitschka (2001) present a first step how business processes and EM/CBR can be
integrated. Further approaches on process-oriented knowledge management and CBR
can be found in Weber and Gresse von Wangenheim (2001). CBR-based knowledge
reuse for project management is described in Althoff, Nick, and Tautz (1999), Tautz
(2000), Brandt and Nick (2001), and Friedrich, Iglezakis et al. (2002). CBR for
supporting knowledge mediation is the topic underlying Griffiths, Harrison, and
Dearden (1999).
4.5
Text Mining in indiGo
Text mining is concerned with the task of extracting relevant information from natural
language text and to search for interesting relationships between the extracted entities.
From a linguistic viewpoint natural language exhibits complex structures on different
hierarchical levels, which are interconnected to each other (H•ebíc•eck 1996). These
structures, however, are tuned to human cognitive abilities. From the perspective of a
computational system, which is adopted here, linguistic information appears to be
implicitly encoded in an unstructured way and presents a challenge for automatic data
processing.
Text classification is one of the basic techniques in the area of text-mining. It
means that text documents are filtered into a set of content-categories. For the task of
