10 Dec Serbian paraphrase corpus: paraphrase.sr

The Ser­bian Para­phrase Cor­pus (paraphrase.sr) con­sists of 1194 pairs of sen­tences gath­ered from news sources on the web. Each sen­tence pair was man­u­al­ly an­no­tat­ed with a bi­na­ry sim­i­lar­i­ty score that in­di­cates whether the sen­tences in the pair are se­man­ti­cal­ly sim­i­lar enough to be con­sid­ered close para­phras­es. The cor­pus con­tains 553 sen­tence pairs deemed to be se­man­ti­cal­ly equiv­a­lent (46.31% of the to­tal num­ber), and 641 se­man­ti­cal­ly di­verse pairs (53.69% of the to­tal num­ber).

• Vuk Batanović, Bo­jan Furlan, Boško Nikolić (2011). A soft­ware sys­tem for de­ter­min­ing the se­man­tic sim­i­lar­i­ty of short texts in Ser­bian. Pro­ceed­ings of the 19th Telecom­mu­ni­ca­tions fo­rum (TELFOR 2011), pp. 1249–1252, Bel­grade, Ser­bia. [Link]
• Bo­jan Furlan, Vuk Batanović, Boško Nikolić (2013). Se­man­tic sim­i­lar­i­ty of short texts in lan­guages with a de­fi­cient nat­ur­al lan­guage pro­cess­ing sup­port. De­ci­sion Sup­port Sys­tems, Vol. 55, No. 3, pp. 710–719. [Link]