Evaluation of Logistics Service Quality: Sentiment Analysis of Comment Text Based on Multi-Level Graph Neural Network
Bibliographic record
Abstract
The world is severely impacted by the coronavirus (COVID19). During the epidemic, logistics service, an often-overlooked pillar of the modern society, steps into the spotlight. However, the service capability is inevitably weakened by the epidemic. The fatigued service providers are increasingly unable to meet the high expectations of users, who therefore leave harsh comments on logistics services. It is important for managers to find information that helps to improve management, out of the biased and angry comments. Text sentiment analysis is a fundamental work in natural language processing (NLP). In recent years, graph neural network (GNN) has achieved excellent performance in various NLP tasks. Nevertheless, GNN only considers the adjacent words, as it updates graph nodes. The model thereby emphasizes local features over global features, and misses the intent of the comment text. This paper constructs a triple graph neural network (TGNN) to serve the sentiment analysis of service texts. Firstly, the corresponding node connection windows were applied on different network layers to consider both local and global features. Next, the graph attention network (GAT) was adopted as the message delivery mechanism to fuse the features of all word nodes in the graph. Experimental results show that, the TGNN can evaluate the comment texts on logistics service quality more accurately than the other models.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".