MétaCan
Menu
Back to cohort
Record W4385461227 · doi:10.21203/rs.3.rs-3208999/v1

Enhanced heterogeneous graph convolutional networks with dual-level attention for aspect-based sentiment analysis

2023· preprint· en· W4385461227 on OpenAlexaff
Haochen Zou, Yongli Wang

Bibliographic record

VenueResearch Square · 2023
Typepreprint
Languageen
FieldComputer Science
TopicSentiment Analysis and Opinion Mining
Canadian institutionsConcordia University
FundersNational Natural Science Foundation of China
KeywordsComputer scienceGraphSentenceDependency (UML)ParsingArtificial intelligenceDependency graphDependency grammarConvolutional neural networkNode (physics)Sentiment analysisNatural language processingDual (grammatical number)Theoretical computer science

Abstract

fetched live from OpenAlex

Abstract Aspect-based sentiment analysis aims to analyze the sentimental tendencies of specific aspect terms in the target sentence. With the continuous development of deep learning technology, graph convolutional networks have been widely applied to sentiment analysis tasks and achieved satisfactory results. However, when constructing graph convolutional networks based on text contents, the model considers the contextual words and their interdependent relationships without distinction. Meanwhile, most models are built separately, without fully employing the information of part of speech and dependency relationships in the text dependency parsing. In response to the above issues, we design a heterogeneous graph neural networks model based on dual node-level and type-level attention mechanisms. The heterogeneous information network is embedded into graph convolutional networks for considering the attribute characteristics of different node types. The dual-level attention mechanism can capture both the importance between diverse adjacent nodes and the significance of different node types to the central node. The relationship between aspect terms and sentiment terms is acquired by adjusting the weights of different types of nodes. To address the issue that graph neural networks have difficulty in fully utilizing dependency parsing information, we embedded syntactic and semantic dependency enhancement. By introducing the dependency tree and the dependency parsing into the graph convolutional networks architecture, it can comprehensively model the syntax relationship and analyze the sentence structure, thereby distinguishing the important contextual information. The experimental results on multiple public datasets indicate the effectiveness of the proposed model in aspect-based sentiment analysis.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.001
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.012
Threshold uncertainty score0.024

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0010.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.112
GPT teacher head0.381
Teacher spread0.269 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2023
Admission routes1
Has abstractyes

Explore more

Same venueResearch SquareSame topicSentiment Analysis and Opinion MiningFrench-language works237,207