Cross-lingual Word Analogies using Linear Transformations between Semantic Spaces
This provides a new intrinsic evaluation method for cross-lingual semantic spaces, which is incremental as it builds on existing linear transformations and dictionaries.
The paper tackles the problem of evaluating cross-lingual semantic spaces by generalizing word analogies across six languages, achieving average accuracies of 51.1% for monolingual, 43.1% for bilingual, and 38.2% for multilingual spaces.
We generalize the word analogy task across languages, to provide a new intrinsic evaluation method for cross-lingual semantic spaces. We experiment with six languages within different language families, including English, German, Spanish, Italian, Czech, and Croatian. State-of-the-art monolingual semantic spaces are transformed into a shared space using dictionaries of word translations. We compare several linear transformations and rank them for experiments with monolingual (no transformation), bilingual (one semantic space is transformed to another), and multilingual (all semantic spaces are transformed onto English space) versions of semantic spaces. We show that tested linear transformations preserve relationships between words (word analogies) and lead to impressive results. We achieve average accuracy of 51.1%, 43.1%, and 38.2% for monolingual, bilingual, and multilingual semantic spaces, respectively.