Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Oxford University Press!

ad

Style, Mediation, and Change

Edited by Janus Mortensen, Nikolas Coupland, and Jacob Thogersen

Style, Mediation, and Change "Offers a coherent view of style as a unifying concept for the sociolinguistics of talking media."


New from Cambridge University Press!

ad

Intonation and Prosodic Structure

By Caroline Féry

Intonation and Prosodic Structure "provides a state-of-the-art survey of intonation and prosodic structure."


The LINGUIST List is dedicated to providing information on language and language analysis, and to providing the discipline of linguistics with the infrastructure necessary to function in the digital world. LINGUIST is a free resource, run by linguistics students and faculty, and supported by your donations. Please support LINGUIST List during the 2017 Fund Drive.

Academic Paper


Title: Evaluating vector space models with canonical correlation analysis
Author: Sami Virpioja
Institution: Aalto University School of Science
Author: Mari-Sanna Paukkeri
Institution: Aalto University School of Science
Author: Abhishek Tripathi
Institution: University of Helsinki
Author: Tiina Lindh-Knuutila
Institution: Aalto University School of Science
Author: Krista Lagus
Institution: Aalto University School of Science
Linguistic Field: Computational Linguistics
Abstract: Vector space models are used in language processing applications for calculating semantic similarities of words or documents. The vector spaces are generated with feature extraction methods for text data. However, evaluation of the feature extraction methods may be difficult. Indirect evaluation in an application is often time-consuming and the results may not generalize to other applications, whereas direct evaluations that measure the amount of captured semantic information usually require human evaluators or annotated data sets. We propose a novel direct evaluation method based on canonical correlation analysis (CCA), the classical method for finding linear relationship between two data sets. In our setting, the two sets are parallel text documents in two languages. A good feature extraction method should provide representations that reflect the semantic content of the documents. Assuming that the underlying semantic content is independent of the language, we can study feature extraction methods that capture the content best by measuring dependence between the representations of a document and its translation. In the case of CCA, the applied measure of dependence is correlation. The evaluation method is based on unsupervised learning, it is language- and domain-independent, and it does not require additional resources besides a parallel corpus. In this paper, we demonstrate the evaluation method on a sentence-aligned parallel corpus. The method is validated by showing that the obtained results with bag-of-words representations are intuitive and agree well with the previous findings. Moreover, we examine the performance of the proposed evaluation method with indirect evaluation methods in simple sentence matching tasks, and a quantitative manual evaluation of word translations. The results of the proposed method correlate well with the results of the indirect and manual evaluations.

CUP AT LINGUIST

This article appears IN Natural Language Engineering Vol. 18, Issue 3.

Return to TOC.

Add a new paper
Return to Academic Papers main page
Return to Directory of Linguists main page