Exploring Semantic Capacity of Terms

Huang, Jie; Wang, Zilong; Chang, Kevin Chen-Chuan; Hwu, Wen-mei W.; Xiong, Jinjun

doi:10.18653/v1/2020.emnlp-main.684

Cited by 7 publications

(20 citation statements)

References 17 publications

(13 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hyperbolic representations have been used to model complex networks (Krioukov et al 2010;Kiela 2017, 2018;Tay, Luu, and Hui 2017;Gülc ¸ehre et al 2018;Huang et al 2020) and have proven more suitable than Euclidean space in representing hierarchical data (Sala et al 2018;Nickel and Kiela 2017). For example, López and Strube (2020) introduce hyperbolic representations to capture latent hierarchies arising from the class distribution for multi-class multi-label classification.…”

Section: Related Workmentioning

confidence: 99%

“…However, there is room for improvement in such seed word based methods. First, they neglect to consider the latent hierarchies between words, and it is assumed that capturing latent hierarchies between words will further improve seed word based methods on aspect inference, for instance by better identifying and organizing seed words and their hypernym pairs (Huang et al 2020;López, Heinzerling, and Strube 2019). For example, as shown in Fig.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hyperbolic Disentangled Representation for Fine-Grained Aspect Extraction

Tai

2022

AAAI

View full text Add to dashboard Cite

Automatic identification of salient aspects from user reviews is especially useful for opinion analysis. There has been significant progress in utilizing weakly supervised approaches, which require only a small set of seed words for training aspect classifiers. However, there is always room for improvement. First, no weakly supervised approaches fully utilize latent hierarchies between words. Second, each seed word’s representation should have different latent semantics and be distinct when it represents a different aspect. In this paper we propose HDAE, a hyperbolic disentangled aspect extractor in which a hyperbolic aspect classifier captures words’ latent hierarchies, and an aspect-disentangled representation models the distinct latent semantics of each seed word. Compared to previous baselines, HDAE achieves average F1 performance gains of 18.2% and 24.1% on Amazon product review and restaurant review datasets, respectively. In addition, the embedding visualization experience demonstrates that HDAE is a more effective approach to leveraging seed words. An ablation study and a case study further attest the effectiveness of the proposed components.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Hyperbolic Disentangled Representation for Fine-Grained Aspect Extraction

Tai

2022

AAAI

View full text Add to dashboard Cite

show abstract

“…In the technological domain, it is ICT driven and computer based. In this regard, semantic analysis entails the process whereby meaning representations are composed and assigned to linguistic inputs (Huang, Wang, Chang, Hwu, & Xiong, 2020;Meštrović, Martinčić-Ipšić, & Čubrilo, 2007). To create rich and accurate meaning representations, a wide range of knowledge-sources and inference techniques are involved.…”

Section: The Nature Of Semantic Analysismentioning

confidence: 99%

A Semantic Analysis of Corona Virus Pandemic Terms

Edet

Babatunde

Ogbulogo

et al. 2021

AJSSMS

View full text Add to dashboard Cite

Various epidemics in recent years have introduced myriad of challenges to the entire world. COVID-19 disease is the latest crisis with its attendant health and language issues. With its emergence, COVID-19 introduced into the global linguistic repertoire an avalanche of unknown vocabulary to the ordinary language user. In today’s world, language is not only available for communication; it is a tool that contributes to maintaining global peace and order. In bridging diverse communities of humanities in the world, shared meaning becomes a platform for mutual understanding, promoting intellectual development and collaborative research efforts. The current study aims to explore and explicate the novel language of COVID-19, thereby making meaning accessible for clarity of communication. The qualitative method of analysis which relied on secondary data from different online COVID-19 glossaries was utilised. Data collection was a total of 149 terms, out of which 34 were purposively selected for analysis. This was to examine their semantic meaning and also ascertain their word relations. The study investigates the process of developing meaning mechanisms in the use COVID-19 terms among language users. The study found that the COVID-19 has a distinct vocabulary that can be analysed linguistically. The literature review in this study highlighted past researches on COVID-19 as descriptive, others on the frequency count of words and the etymology of terms.

show abstract

“…To overcome the limitations mentioned above, we develop an exploration system that manages corpus sentences as a descriptive knowledge graph (Huang et al, 2022b). The Descriptive knowledge graph for Explaining Entity Relationships (DEER) is a special knowledge graph where each edge is not a relation label but a set of relational sentences describing the relationship between a pair of entities (Handler and O'Connor, 2018;Huang et al, 2022a;Huang and Chang, 2022;Liu et al, 2023;.…”

Section: Introductionmentioning

confidence: 99%

Descriptive Knowledge Graph in Biomedical Domain

Zhu,

Huang,

Chang

2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: System Demonstrations

View full text Add to dashboard Cite

We present a novel system that automatically extracts and generates informative and descriptive sentences from the biomedical corpus and facilitates the efficient search for relational knowledge. Unlike previous search engines or exploration systems that retrieve unconnected passages, our system organizes descriptive sentences as a relational graph, enabling researchers to explore closely related biomedical entities (e.g., diseases treated by a chemical) or indirectly connected entities (e.g., potential drugs for treating a disease). Our system also uses ChatGPT and a fine-tuned relation synthesis model to generate concise and reliable descriptive sentences from retrieved information, reducing the need for extensive human reading effort. With our system, researchers can easily obtain both high-level knowledge and detailed references and interactively steer to the information of interest. We spotlight the application of our system in COVID-19 research, illustrating its utility in areas such as drug repurposing and literature curation. 1

show abstract

Exploring Semantic Capacity of Terms

Cited by 7 publications

References 17 publications

Hyperbolic Disentangled Representation for Fine-Grained Aspect Extraction

Hyperbolic Disentangled Representation for Fine-Grained Aspect Extraction

A Semantic Analysis of Corona Virus Pandemic Terms

Descriptive Knowledge Graph in Biomedical Domain

Contact Info

Product

Resources

About