English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Though there are some works on improving distributed word representationsusing lexicons, the improper overfitting of the words that have multiplemeanings is a remaining issue deteriorating the learning when lexicons areused, which needs to be solved. An alternative method is to allocate a vectorper sense instead of a vector per word. However, the word representationsestimated in the former way are not as easy to use as the latter one. Ourprevious work uses a probabilistic method to alleviate the overfitting, but itis not robust with a small corpus. In this paper, we propose a new neuralnetwork to estimate distributed word representations using a lexicon and acorpus. We add a lexicon layer in the continuous bag-of-words model and athreshold node after the output of the lexicon layer. The threshold rejects theunreliable outputs of the lexicon layer that are less likely to be the samewith their inputs. In this way, it alleviates the overfitting of the polysemouswords. The proposed neural network can be trained using negative sampling,which maximizing the log probabilities of target words given the context words,by distinguishing the target words from random noises. We compare the proposedneural network with the continuous bag-of-words model, the other worksimproving it, and the previous works estimating distributed wordrepresentations using both a lexicon and a corpus. The experimental resultsshow that the proposed neural network is more efficient and balanced for bothsemantic tasks and syntactic tasks than the previous works, and robust to thesize of the corpus.

Alleviating Overfitting for Polysemous Words for Word Representation Estimation Using Lexicons