"... In PAGE 4: ... Excluding a single metonymic name from the index means that the in-term-frequency is lowered and not necessarily zero. The results shown in Table5 indicate that performance is improved when querying the index containing only the lit- eral senses for location names. Computation of recall values is based on official relevance assessments of documents.... ..."

"... In PAGE 7: ... We intend (time permitting) to rerun a number of the queries to see if, and how, these errors a ected the results. 4 Results for Submitted Runs The summary results (as Mean Average Precision) for the submitted bilingual and monolingual runs for both English and German are shown in Table2 , the Recall-Precision curves for these runs are also shown in Figures 1 (for monolingual) and 2 (for bilingual). In Figures 1 and 2 the name are abbrevated to the nal letters and numbers of the full name in Table 2, and those beginning with \POST quot; are uno cial runs described in the next section.... In PAGE 7: ... 4 Results for Submitted Runs The summary results (as Mean Average Precision) for the submitted bilingual and monolingual runs for both English and German are shown in Table 2, the Recall-Precision curves for these runs are also shown in Figures 1 (for monolingual) and 2 (for bilingual). In Figures 1 and 2 the name are abbrevated to the nal letters and numbers of the full name in Table2 , and those beginning with \POST quot; are uno cial runs described in the next section. Table 2 indicates some rather curious results that warrant further investigation as to the cause.... In PAGE 7: ... In Figures 1 and 2 the name are abbrevated to the nal letters and numbers of the full name in Table 2, and those beginning with \POST quot; are uno cial runs described in the next section. Table2 indicates some rather curious results that warrant further investigation as to the cause. Notice that the result for all of the English monolingual runs exceed the rsults for bilingual German to English runs - this is typical for cross-langauge retrieval.... In PAGE 8: ...3879 POSTMLENNOL Monolingual English no 0.3615 Table 3: Additional Post-Submission GeoCLEF Runs As can be seen by comparing Table 3 with Table2 , all of the comparable runs for show... ..."

"... In PAGE 5: ... query fields: combinations of title (T), description (D), and locations from narrative (N). Parameters and results for monolingual German and bilingual English-German experiments are shown in Table1 . The table shows relevant and retrieved documents (rel ret), MAP and precision at five, ten, and twenty documents.... ..."

"... In PAGE 6: ... The GeoCLEF 2005 topics and relevance judgments are used as training data. 4 Results Table3 summarises the results of our runs using evaluation metrics include Average Precision, R-Precision and the increment over the mean average precision (i.e.... ..."

"... In PAGE 6: ...Table1 lists the indexes created for the GeoCLEF database and the document elements from which the contents of those indexes were extracted. The \Used quot; column in Table 1 indicates whether or not a particular index was used in the submitted GeoCLEF runs.... In PAGE 6: ...which the contents of those indexes were extracted. The \Used quot; column in Table1 indicates whether or not a particular index was used in the submitted GeoCLEF runs. Because there was no explicit tagging of location-related terms in the collections used for GeoCLEF, we applied the above approach to the \TEXT quot;, \LD quot;, and \TX quot; elements of the records of the various collections.... In PAGE 6: ... For example when running search against the English databases the name \Europe quot; was expanded to \Albania Andorra Austria Belarus Bel- gium Bosnia and Herzegovina Bulgaria Croatia Cyprus Czech Republic Denmark Estonia Faroe Islands Finland France Georgia Germany Gibraltar Greece Guernsey and Alderney Hungary Ice- land Ireland Isle of Man Italy Jersey Latvia Liechtenstein Lithuania Luxembourg Macedonia Malta Moldova Monaco Netherlands Norway Poland Portugal Romania Russia San Marino Serbia and Montenegro Slovakia Slovenia Spain Svalbard and Jan Mayen Sweden Switzerland Turkey Ukraine United Kingdom Vatican City quot;, while for searches against the German databases \Europa quot; was expanded to \Albanien Andorra sterreich Weirussland Belgien Bosnien und Herzegowina Bulgar- ien Kroatien Zypern Tschechische Republik Dnemark Estland Frer-Inseln Finnland Frankreich Georgien Deutschland Gibraltar Griechenland Guernsey und Alderney Ungarn Island Irland Man Italien Jersey Lettland Liechtenstein Litauen Luxemburg Mazedonien Malta Moldawien Monaco Niederlande Norwegen Polen Portugal Rumnien Russland San Marino Serbien und Montenegro Slowakei Slowenien Spanien Svalbard und Jan Mayen Schweden Schweiz Trkei Ukraine Grobri- tannien Vatikan quot;. Example queries for monolingual searches are shown in Figure 3 The indexes combined in searching included the headline, topic, and geotext indexes (as de- scribed in Table1 ) for searches that include the location element, and the headline and topic for the searches without the locations element. For the bilingual tasks, three sub-queries, one for each query translation were run and then the results were merged using the CombMNZ algorithm.... ..."

"... In PAGE 12: ...62% and a Precision at 10 documents of just 36%. Table3 gives a summary of the averaged results for each run. As for the terminology, all run names start the letter L followed by an indicator of how the query was formed.... ..."

"... In PAGE 2: ...escription plus concept-location is about 5% better (0.3936 versus 0.3737). The blind feedback results for English are summarized in Tables 1a (monolingual) and 1b (bilingual German English): Table1 a: Berkeley2 GeoCLEF English Monolingual Run Name Type MAP- blind feedback (BF) MAP-no BF BKGeoE1 TD+Concept/Locat. (CL) 0.... ..."

"... In PAGE 4: ... Table2 lists the indexes created by the Cheshire II system for the GeoCLEF database and the document elements from which the contents of those indexes were extracted. The Used column in Table 2 indicates whether or not a particular index was used in the submitted GeoCLEF runs.... In PAGE 4: ...Table 2 lists the indexes created by the Cheshire II system for the GeoCLEF database and the document elements from which the contents of those indexes were extracted. The Used column in Table2 indicates whether or not a particular index was used in the submitted GeoCLEF runs. The georeferencing indexing subsystem of Cheshire II was used for the geotext, geopoint, and geobox indexes.... ..."

"... In PAGE 3: ... A run with topic title, description, and narrative was mandatory to find out if extra narrative terms help improve performance. Table1 and Table 2 show the different parameter settings and results for our monolingual German and bilingual English-German GeoCLEF runs, respectively. The results shown consist of mean average precision (MAP) and the number of relevant and retrieved documents (rel ret).... ..."

"... In PAGE 3: ... A run with topic title, description, and narrative was mandatory to find out if extra narrative terms help improve performance. Table 1 and Table2 show the different parameter settings and results for our monolingual German and bilingual English-German GeoCLEF runs, respectively. The results shown consist of mean average precision (MAP) and the number of relevant and retrieved documents (rel ret).... ..."