The KBS-CROSS Project

KBS-Media Lab
Lund University


[projects] Last update 1995.05.29

Knowledge-Based System for automatic CROSS-referencing of classification systems - KBS-CROSS

Goal/Purpose:

The goals of KBS-CROSS are:

The intended applications of KBS-CROSS was:

The project has to a higher extent than inticipated been forced to focus on the data handling and transformation process due to system and handling routines incompatibilities.

Relevance:

Mainly relevant for research and education, and hence the funding comes from FRN for research in library science.

Novelty:

Neural networks have been deployed in many different ways in information retrieval (Scholtes), (Belew), (NIST). Cross-referencing classification systems with neural networks though, has to our knowledge not been done previously.

Methodology:

Data from the LIBRIS database is used together with neural network and text processing tools. The resulting code and data is deployed in a demonstration tool in a Dynamic Knowledge Net. The tool's performance is continuously evaluated by indexing specialists.

Results:

The output of KBS-CROSS is a prototype that cross-references LCC/LSH with UDC and a report on the project. The report will be in English. One or more article proposals will also be produced. Special emphasis will be on the internal library data handling routines and recommendations for the future.

Project set-up:


Feedback Navigate Front Page [projects]