Sinhala Sentence Embedding: A Two-Tiered Structure for Low-Resource Languages
This work addresses the problem of limited resources for Sinhala language processing, offering an incremental improvement in embedding methods for sentiment analysis.
The paper tackled the challenge of developing functional sentence embeddings for the low-resource Sinhala language by comparing one-tiered and two-tiered architectures, finding that a two-tiered approach with word and sentence embeddings achieved a maximum F1 score of 88.04%, outperforming word embeddings at 83.76% in sentiment analysis.
In the process of numerically modeling natural languages, developing language embeddings is a vital step. However, it is challenging to develop functional embeddings for resource-poor languages such as Sinhala, for which sufficiently large corpora, effective language parsers, and any other required resources are difficult to find. In such conditions, the exploitation of existing models to come up with an efficacious embedding methodology to numerically represent text could be quite fruitful. This paper explores the effectivity of several one-tiered and two-tiered embedding architectures in representing Sinhala text in the sentiment analysis domain. With our findings, the two-tiered embedding architecture where the lower-tier consists of a word embedding and the upper-tier consists of a sentence embedding has been proven to perform better than one-tier word embeddings, by achieving a maximum F1 score of 88.04% in contrast to the 83.76% achieved by word embedding models. Furthermore, embeddings in the hyperbolic space are also developed and compared with Euclidean embeddings in terms of performance. A sentiment data set consisting of Facebook posts and associated reactions have been used for this research. To effectively compare the performance of different embedding systems, the same deep neural network structure has been trained on sentiment data with each of the embedding systems used to encode the text associated.