Gabriella Lapesa

Contrasting Syntagmatic and Paradigmatic Relations: Insights from Distributional Semantic Models

This paper presents a large-scale evaluation of bag-of-words distributional models on two datasets from priming experiments involving syntagmatic and paradigmatic relations. We interpret the variation in performance achieved by different settings of the model parameters as an indication of which aspects of distributional patterns characterize these types of relations. Contrary to what has been argued in the literature (Rapp, 2002; Sahlgren, 2006) ‐ that bag-of-words models based on secondorder statistics mainly capture paradigmatic relations and that syntagmatic relations need to be gathered from first-order models ‐ we show that second-order models perform well on both paradigmatic and syntagmatic relations if their parameters are properly tuned. In particular, our results show that size of the context window and dimensionality reduction play a key role in differentiating DSM performance on paradigmatic vs. syntagmatic relations.

SemantiKLUE: Semantic Textual Similarity with Maximum Weight Matching

This paper describes the SemantiKLUE system (Proisl et al., 2014) used for the SemEval2015 shared task on Semantic Textual Similarity (STS) for English. The system was developed for SemEval-2013 and extended for SemEval-2014, where it participated in three tasks and ranked 13th out of 38 submissions for the English STS task. While this year’s submission ranks 46th out of 73, further experiments on the selection of training data led to notable improvements showing that the system could have achieved rank 22 out of 73. We report a detailed analysis of those training selection experiments in which we tested different combinations of all the available STS datasets, as well as results of a qualitative analysis conducted on a sample of the sentence pairs for which SemantiKLUE gave wrong STS predictions.

NaDiR: Naive Distributional Response Generation

This paper describes NaDiR (Naive DIstributional Response generation), a corpus-based system that, from a set of word stimuli as an input, generates a response word relying on association strength and distributional similarity. NaDiR participated in the CogALex 2014 shared task on multiword associations (restricted systems track), operationalizing the task as a ranking problem: candidate words from a large vocabulary are ranked by their average association or similarity to a given set of stimuli. We also report on a number of experiments conducted on the shared task data, comparing first-order models (based on co-occurrence and statistical association) to second-order models (based on distributional similarity).