Technical Meeting

21 January 14h-18h

Crawling/indexing (1 1/2 hours)

• architecture: interactions between crawler, repository, text extraction, morphological and semantic analysis, indexer
• XML format for results of morphological and semantic analysis
• increase size of collection
• better detection of educational research documents (includes ISN classifier integration)
• more metadata (what metadata is important, e.g. date published, title, author)

Break (30 min)

Querying (other than multilingual functionality) (1 hour)

• use of key sentences for ranking and displaying results
• interface usability and additional features
• transfer of web interface development to Xerox?

Evaluation of semantic extraction, multilingual functionality (1 hour)

22. January 13h-15h
two-hour technical session in parallel with semantic session

WP9 multilingual functionality (1 1/2 hours, divided roughly as shown)

• discuss resource types and how to use them in search engine (45 minutes)
• IRDP & DIPF present their work on term networks
• general-purpose resources
• bilingual dictionaries
• (monolingual thesauri and subject headings without concordances: are they useful?)
• morphological analyzers
• share info on available resources for 15 languages (15 min)
• query translation user interface (1/2 hour)
• look at CACAO interface and RRZN thesaurus demo together

Wrap-up (1/2 hour)