"... In PAGE 3: ... 4 Data for Partial Cognates We performed experiments with ten pairs of par- tial cognates. We list them in Table1 . For a French partial cognate we list its English cognate and several false friends in English.... In PAGE 3: ... All partial cognates are words frequently used in the language. We selected ten partial cognates pre- sented in Table1 according to the number of ex- tracted sentences (a balance between the two meanings), to evaluate and experiment our pro- posed methods. The human effort that we required for our methods was to add more false-friend English words, than the ones we found in the web re- source.... ..."

"... In PAGE 83: ...ECIDE, WP1B, sec.6.5.3 Features WordCruncher COSMAS PROSA Church ISI xlex TACT simple index X X ? X { X word/pattern frequency X X ? X X X KWIC{ context spec. * # characters X { X X X { * # word forms { X X X X X * text structure (X) X X { { X show citation ? X ? { { X sorting{ position in KWIC line X X X (X) { X { complex criteria { { X { { { { sort order control X X X (X) { X display tag values { { { { { X display aligned citation ? { { X { { store results in a le X X X X X X Table 19: Presentation of query results III Features WordCruncher COSMAS PROSA Church ISI xlex TACT grouping of concordance lines { manually { X { { { { { automatically { { { { { { reduction of concordance lines { manually { X { { { { { automatically * random selection { X { { { { * other { { { { { { { use as new subcorpus { X { { { { built-in statistics { MI { { { X { { { t{score { { { X { { { Z{score X { { { { X { other { { X { X { Table2 0: Postprocessing of query results III Features WordCruncher COSMAS PROSA Church ISI xlex TACT plain text X X X X X X tagged text { { { { { (X) SGML encoded text { { { { { { bilingual text (X) { { X { { search is performed on { textual data { { X { (X) { { special encoded data X X { X X X distinction between text amp; tags { in data representation { { { { { { { in query system X { { { { X Table 21: Input format of corpora III Ulrich Heid, Bruno M. Schulze (STR) 78... In PAGE 83: ...ECIDE, WP1B, sec.6.5.3 Features WordCruncher COSMAS PROSA Church ISI xlex TACT simple index X X ? X { X word/pattern frequency X X ? X X X KWIC{ context spec. * # characters X { X X X { * # word forms { X X X X X * text structure (X) X X { { X show citation ? X ? { { X sorting{ position in KWIC line X X X (X) { X { complex criteria { { X { { { { sort order control X X X (X) { X display tag values { { { { { X display aligned citation ? { { X { { store results in a le X X X X X X Table 19: Presentation of query results III Features WordCruncher COSMAS PROSA Church ISI xlex TACT grouping of concordance lines { manually { X { { { { { automatically { { { { { { reduction of concordance lines { manually { X { { { { { automatically * random selection { X { { { { * other { { { { { { { use as new subcorpus { X { { { { built-in statistics { MI { { { X { { { t{score { { { X { { { Z{score X { { { { X { other { { X { X { Table 20: Postprocessing of query results III Features WordCruncher COSMAS PROSA Church ISI xlex TACT plain text X X X X X X tagged text { { { { { (X) SGML encoded text { { { { { { bilingual text (X) { { X { { search is performed on { textual data { { X { (X) { { special encoded data X X { X X X distinction between text amp; tags { in data representation { { { { { { { in query system X { { { { X Table2 1: Input format of corpora III Ulrich Heid, Bruno M. Schulze (STR) 78... In PAGE 84: ...ECIDE, WP1B, sec.6.5.3 Features WordCruncher COSMAS PROSA Church ISI xlex TACT usage mode { interactive * menus X X X { { X * graphical user interface { { { { X { * query language { { { X { X { batch { { ? X ? X Table2 2: User modes III Features WordCruncher COSMAS PROSA Church ISI xlex TACT max. size of corpora not restricted X ? X X ? { support for char sets of DE, FR, EN X { X { ? X integratable into DECIDE toolbox { { { X ? X application test successful { { { { { { Table 23: Miscellaneous III Ulrich Heid, Bruno M.... In PAGE 84: ...ECIDE, WP1B, sec.6.5.3 Features WordCruncher COSMAS PROSA Church ISI xlex TACT usage mode { interactive * menus X X X { { X * graphical user interface { { { { X { * query language { { { X { X { batch { { ? X ? X Table 22: User modes III Features WordCruncher COSMAS PROSA Church ISI xlex TACT max. size of corpora not restricted X ? X X ? { support for char sets of DE, FR, EN X { X { ? X integratable into DECIDE toolbox { { { X ? X application test successful { { { { { { Table2 3: Miscellaneous III Ulrich Heid, Bruno M. Schulze (STR) 79... In PAGE 86: ... { { { { { word length { { { { { external tools { { { { search space spec. { # characters { { X { { # word forms X { { X { text structure ? { X { query method { re nement of search conditions (X) X (X) X { subqueries { { { { { combination of query results { { { { Table2 4: Search capabilities IV Ulrich Heid, Bruno M. Schulze (STR) 81... In PAGE 87: ...ECIDE, WP1B, sec.6.5.4 Features CoCo LQTEXT LEXA Kaye built-in corpora X { { { own corpora ? X X X de nition of subsets { (X) { { Table2 5: Selection of corpora IV Features CoCo LQTEXT LEXA Kaye simple index { { X { word/pattern frequency ? { X X KWIC{ context spec. * # characters { X ? { * # word forms { { ? { * text structure (X) (X) ? { show citation { X X X sorting{ position in KWIC line { { X (X) { complex criteria { { { { { sort order control { { X { display tag values X { { X display aligned citation { { { X store results in a le (X) X X X Table 26: Presentation of query results IV Features CoCo LQTEXT LEXA Kaye grouping of concordance lines { manually { { { { { automatically { { { { reduction of concordance lines { manually X { ? { { automatically * random selection { { ? { * other { { ? { { use as new subcorpus { { { { built-in statistics { MI X { { { { t{score { { { { { Z{score X { { { { other X { { { Table 27: Postprocessing of query results IV Ulrich Heid, Bruno M.... In PAGE 87: ...ECIDE, WP1B, sec.6.5.4 Features CoCo LQTEXT LEXA Kaye built-in corpora X { { { own corpora ? X X X de nition of subsets { (X) { { Table 25: Selection of corpora IV Features CoCo LQTEXT LEXA Kaye simple index { { X { word/pattern frequency ? { X X KWIC{ context spec. * # characters { X ? { * # word forms { { ? { * text structure (X) (X) ? { show citation { X X X sorting{ position in KWIC line { { X (X) { complex criteria { { { { { sort order control { { X { display tag values X { { X display aligned citation { { { X store results in a le (X) X X X Table2 6: Presentation of query results IV Features CoCo LQTEXT LEXA Kaye grouping of concordance lines { manually { { { { { automatically { { { { reduction of concordance lines { manually X { ? { { automatically * random selection { { ? { * other { { ? { { use as new subcorpus { { { { built-in statistics { MI X { { { { t{score { { { { { Z{score X { { { { other X { { { Table 27: Postprocessing of query results IV Ulrich Heid, Bruno M. Schulze (STR) 82... In PAGE 87: ...ECIDE, WP1B, sec.6.5.4 Features CoCo LQTEXT LEXA Kaye built-in corpora X { { { own corpora ? X X X de nition of subsets { (X) { { Table 25: Selection of corpora IV Features CoCo LQTEXT LEXA Kaye simple index { { X { word/pattern frequency ? { X X KWIC{ context spec. * # characters { X ? { * # word forms { { ? { * text structure (X) (X) ? { show citation { X X X sorting{ position in KWIC line { { X (X) { complex criteria { { { { { sort order control { { X { display tag values X { { X display aligned citation { { { X store results in a le (X) X X X Table 26: Presentation of query results IV Features CoCo LQTEXT LEXA Kaye grouping of concordance lines { manually { { { { { automatically { { { { reduction of concordance lines { manually X { ? { { automatically * random selection { { ? { * other { { ? { { use as new subcorpus { { { { built-in statistics { MI X { { { { t{score { { { { { Z{score X { { { { other X { { { Table2 7: Postprocessing of query results IV Ulrich Heid, Bruno M. Schulze (STR) 82... In PAGE 88: ...ECIDE, WP1B, sec.6.5.4 Features CoCo LQTEXT LEXA Kaye plain text X X X X tagged text X { { X SGML encoded text { { { { bilingual text { { { X search is performed on { textual data X { X { { special encoded data { X X X distinction between text amp; tags { in data representation { { { X { in query system { { { { Table2 8: Input format of corpora IV Features CoCo LQTEXT LEXA Kaye usage mode { interactive * menus X X ? X * graphical user interface X { { { * query language { X X { { batch { X X { Table 29: User modes IV Features CoCo LQTEXT LEXA Kaye max. size of corpora not restricted { { { { support for char sets of DE, FR, EN X ? ? X integratable into DECIDE toolbox { X X { application test successful { { { { Table 30: Miscellaneous IV Ulrich Heid, Bruno M.... In PAGE 88: ...ECIDE, WP1B, sec.6.5.4 Features CoCo LQTEXT LEXA Kaye plain text X X X X tagged text X { { X SGML encoded text { { { { bilingual text { { { X search is performed on { textual data X { X { { special encoded data { X X X distinction between text amp; tags { in data representation { { { X { in query system { { { { Table 28: Input format of corpora IV Features CoCo LQTEXT LEXA Kaye usage mode { interactive * menus X X ? X * graphical user interface X { { { * query language { X X { { batch { X X { Table2 9: User modes IV Features CoCo LQTEXT LEXA Kaye max. size of corpora not restricted { { { { support for char sets of DE, FR, EN X ? ? X integratable into DECIDE toolbox { X X { application test successful { { { { Table 30: Miscellaneous IV Ulrich Heid, Bruno M.... ..."

"... In PAGE 4: ...), when sending messages (finding receivers agent location) or when periodical updates are sent to that central point, excess traffic is generated. Since sending large messages also decreases the total number of messages ( Table3 . where SL represents the SLCodec and Java the Java serialization mechanism), this excess traffic has a greater impact on the total overhead.... ..."