Sentiment Analysis Technologies of Advertising Images Based on Deep Learning
Bibliographic record
Abstract
With the rapid advancement of the digital economy, advertising images have emerged as a core medium for brand communication.Sentiment classification of such images plays a critical role in precision marketing and the enhancement of user experience.Recent progress in deep learning offers novel approaches to sentiment analysis in advertising images.However, existing methods remain constrained by limitations in multimodal information fusion, multi-granularity visual feature processing, and attribute-level sentiment interpretation.Common issues include the simplistic concatenation of visual and textual features, neglect of the emotional correlation between local and global visual elements, and the absence of effective integration of multi-scale attribute-level perspectives, all of which result in inadequate classification accuracy and robustness.To address these challenges, an attribute-level sentiment analysis model based on multi-level vision-language alignment and fusion for advertising images was proposed.Through a multi-granularity visual information alignment technique, the model enables precise semantic matching between visual elements-at the pixel, region, and object levels-and their corresponding textual counterparts.Furthermore, multi-scale attribute-level viewpoints were integrated to capture emotional features across dimensions such as color, shape, and embedded textual content.A text-centered multimodal training strategy was also designed to filter irrelevant visual noise.Experimental results demonstrate significant improvements in both accuracy and robustness of advertising image sentiment classification.This model provides technical support for advertising effectiveness evaluation and strategy optimization.The findings contribute theoretically to the advancement of multimodal sentiment analysis and offer practical guidance for precision advertising in contexts such as e-commerce and social media.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".