Retrospective Document Conversion Application to the Library Domain

This paper describes a framework for retrospective document conversion in the library domain. Drawing on the experience and insight gained from projects launched over the present decade by the European Commission, it outlines the requirements for solving the problem of retroconversion and traces the main phases of associated processing. To highlight the main problems encountered in this area, the paper also outlines studies conducted by our group in the more project for the retroconversion of old catalogues belonging to two diierent Libraries : National French Library and Royal Belgian Library. F or the French Library, the idea was to study the feasibility of a recognition approach a voiding the use of ocr and basing the strategy mainly on visual features. The challenge was to recognize a logical structure from its physical aspects. The modest results obtained from experiments for this rst study led us, in the second study, to base the structural recognition methodology more on the logical aspects by focussing the analysis on the content. Furthermore, for the Bel-gian references, the aim was to convert reference catalogues into a more conventional unimarc format while respecting the industrial constraints.Without manual intervention , 75% rate of correct recognition was obtained on 11 catalogues containing about 4548 references. The success of library automation,resulting in user-friendly on-line catalogues 1 integrated with the web and other circulation-systems facilities, has created an urgent need for retroconversion of the older parts of catalogues 2,14,28,31]. As users get familiar with the new catalogue medium, the documents not registered in machine readable form become \invisible" and unreadable. 1 A catalogue is a list of bibliographic descriptions of works. This has meant for many libraries the relegation of an important part of their rich s t o c k of documents to a state of inaccessibility. Such o b vious waste of library collections in addition to the cost diierence between manual handling and an equivalent set of automatic routines has made a strong case for the need to convert a library's entire collection of works to machine-readable records, in the interest of ensuring an eecient use of the investment in the new technology. This has led to the search for cost-eeective tools for the conversion of old catalogues into machine-readable forms. This search has not been limited to the sole problem of conversion but has been extended to embracing other objectives such as ensuring very high rates of …