Snajder, J.; Almic, P.: Modeling semantic compositionality of Croatian multiword expressions (2015)
0.02
0.0209871 = product of:
0.0629613 = sum of:
0.0629613 = product of:
0.09444195 = sum of:
0.058156405 = weight(_text_:network in 2920) [ClassicSimilarity], result of:
0.058156405 = score(doc=2920,freq=2.0), product of:
0.19699359 = queryWeight, product of:
4.4533744 = idf(docFreq=1398, maxDocs=44218)
0.04423468 = queryNorm
0.29521978 = fieldWeight in 2920, product of:
1.4142135 = tf(freq=2.0), with freq of:
2.0 = termFreq=2.0
4.4533744 = idf(docFreq=1398, maxDocs=44218)
0.046875 = fieldNorm(doc=2920)
0.036285542 = weight(_text_:29 in 2920) [ClassicSimilarity], result of:
0.036285542 = score(doc=2920,freq=2.0), product of:
0.15560377 = queryWeight, product of:
3.5176873 = idf(docFreq=3565, maxDocs=44218)
0.04423468 = queryNorm
0.23319192 = fieldWeight in 2920, product of:
1.4142135 = tf(freq=2.0), with freq of:
2.0 = termFreq=2.0
3.5176873 = idf(docFreq=3565, maxDocs=44218)
0.046875 = fieldNorm(doc=2920)
0.6666667 = coord(2/3)
0.33333334 = coord(1/3)
- Abstract
- A distinguishing feature of many multiword expressions (MWEs) is their semantic non-compositionality. Determining the semantic compositionality of MWEs is important for many natural language processing tasks. We address the task of modeling semantic compositionality of Croatian MWEs. We adopt a composition-based approach within the distributional semantics framework. We build and evaluate models based on Latent Semantic Analysis and the recently proposed neural network-based Skip-gram model, and experiment with different composition functions. We show that the compositionality scores predicted by the Skip-gram additive models correlate well with human judgments (=0.50). When framed as a classification task, the model achieves an accuracy of 0.64.
- Date
- 29. 4.2016 12:42:17