Integrating New Languages in a Multilingual Search System Based on a Deep Linguistic Analysis

The LIC2M has designed a cross-lingual search engine based on a deep linguistic analysis of documents and queries that works on French, English, Spanish, German, Arabic and Chinese. For our participation in the CLEF 2004 campaign, we tested the integration in our system of Russian and Finnish, based on a simplified processing. The results we obtained are not good on the new languages introduced, which shows that our system strongly depends on a correct linguistic analysis of the documents. However, integrating more processing steps in the simplified analysis of new languages so that the results of this analysis are more comparable with the results of the complete linguistic analysis seems to be a good direction for improvements.