Papers by John P. McCrae

12 papers
Linghub2: Language Resource Discovery Tool for Language Technologies (2022.lrec-1)

Copied to clipboard

Challenge: Linghub is a platform for language resources that can be used to find and retrieve data . the platform is based on a popular open source data management system, DSpace .
Approach: This work describes a rejuvenation and modernisation of the 2015 platform into using a popular open source data management system, DSpace, as foundation.
Outcome: Linghub2 1 aims to help language resources and technology users find and retrieve relevant data . the new platform, Ling hub2, contains updated and extended resources and more languages offered .
Teanga: A Linked Data based platform for Natural Language Processing (L18-1)

Copied to clipboard

Challenge: Using linked data, we can use many NLP services from a single interface . integrating components within a development model is endemic to software development .
Approach: They propose a linked data based platform for natural language processing that uses linked data to define the types of services input and output.
Outcome: The proposed platform is easy to install and run, easy to use and able to run multiple NLP tasks from one interface.
Cross-lingual Sentence Embedding using Multi-Task Learning (2021.emnlp-main)

Copied to clipboard

Challenge: Existing multilingual sentence embedding models require large parallel corpora to learn efficiently, limiting their scope.
Approach: They propose a sentence embedding framework based on an unsupervised loss function . they capture semantic similarity and relatedness between sentences using a multi-task loss function.
Outcome: The proposed framework outperforms state-of-the-art methods on STS, BUCC and Tatoeba benchmarks and on a monolingual benchmark.
Contextual Modulation for Relation-Level Metaphor Identification (2020.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to identifying metaphors in text ignore context where metaphor occurs . existing approaches focus on word-level identification without explicitly modelling interaction between metaphor components .
Approach: They propose a method for identifying relation-level metaphoric expressions of certain grammatical relations based on contextual modulation.
Outcome: The proposed architecture achieves state-of-the-art results on benchmark datasets.
Unsupervised Deep Language and Dialect Identification for Short Texts (2020.coling-main)

Copied to clipboard

Challenge: Existing methods for identifying closely related short texts are unsupervised . however, performance is poor for unsupervised methods for short texts .
Approach: They propose a method which can learn sentence embeddings and cluster assignments from short texts.
Outcome: The proposed method outperforms state-of-the-art methods in supervised settings . it can learn sentence embeddings and cluster assignments from short texts .
A supervised approach to taxonomy extraction using word embeddings (L18-1)

Copied to clipboard

Challenge: a recent evaluation of a method for organizing texts into a hierarchy showed that it did not outperform a baseline.
Approach: They propose a method that uses supervised learning to combine multiple features with a support vector machine classifier including the baseline features.
Outcome: The proposed method outperforms the baseline method and provides stronger method for identifying taxonomic relations than previous methods.
Suggest me a movie for tonight: Leveraging Knowledge Graphs for Conversational Recommendation (2020.coling-main)

Copied to clipboard

Challenge: Recent studies show that knowledge graphs are incomplete since they do not contain all factual information present on the web.
Approach: They propose to use knowledge graphs to improve the performance of conversational recommender systems by incorporating pre-trained embeddings from subgraphs and positional embeddments into their models.
Outcome: The proposed method improves by 5.62% over the state-of-the-art method on multiple metrics on the recommendation task.
Towards the Construction of a WordNet for Old English (2022.lrec-1)

Copied to clipboard

Challenge: In this paper we discuss our preliminary work towards the construction of a WordNet for Old English, taking our inspiration from other similar WN construction projects for ancient languages such as Ancient Greek, Latin and Sanskrit.
Approach: They propose to use a legacy Old English dictionary to build a WordNet for Old English using a lexicographic resource and the naisc system to automatically compile a provisional version of the WordNet.
Outcome: The proposed OldEWN will be based on lemmas and definitions extracted from a legacy Old English dictionary and will be automatically compile and enriched by experts using the naisc system.
MHE: Code-Mixed Corpora for Similar Language Identification (2022.lrec-1)

Copied to clipboard

Challenge: a new corpus of code-mixed data-sets is presented for similar language identification . the data-settings are based on a more-resourced minority language, Magahi .
Approach: They propose a Magahi-Hindi-English code-mixed corpus for similar language identification . they discuss the complexity of the data-set and provide a few baselines .
Outcome: The proposed corpus provides a language id at two levels: word and sentence.
MaCmS: Magahi Code-mixed Dataset for Sentiment Analysis (2024.lrec-main)

Copied to clipboard

Challenge: Sociolinguists and psychologists have been studying these variations in the lexicons and the language from the 50's . code-mixing is a popular method for understanding people's emotions and attitudes towards various subjects, but low-resourced languages often have a mix of scripts and languages.
Approach: They introduce a new sentiment data, MaCMS, for Magahi-Hindi-English code-mixed language, where Magai is a less-resourced minority language.
Outcome: The proposed dataset is the first Magahi-Hindi-English code-mixed dataset for sentiment analysis tasks.
Automatic Enrichment of Terminological Resources: the IATE RDF Example (L18-1)

Copied to clipboard

Challenge: a recent paper aims to automate the maintenance of terminological resources.
Approach: They propose automatic approaches to maintain and increase lexical coverage of knowledge bases by using machine translation and multilingual word sense disambiguation.
Outcome: The proposed approach outperforms the existing methods with random sentences in most languages .
A Comparison Of Emotion Annotation Schemes And A New Annotated Data Set (L18-1)

Copied to clipboard

Challenge: a series of study on positive/negative sentiments has been conducted on tweets, but recognition of more nuanced affect has received little attention . valence, arousal, dominance and surprise are the most commonly used emotion representation schemes .
Approach: They propose to annotate tweets with scores on four emotion dimensions . they compare annotator agreement with relative annotation schemes over categorical ones .
Outcome: The proposed model improves agreement with relative annotation schemes over categorical ones on Ekman's six basic emotions.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations