Multi-sentiment Modeling with Scalable Systematic Labeled Data Generation via Word2Vec Clustering
Bibliographic record
Abstract
Social networks are now a primary source for news and opinions on topics ranging from sports to politics. Analyzing opinions with an associated sentiment is crucial to the success of any campaign (product, marketing, or political). However, there are two significant challenges that need to be overcome. First, social networks produce large volumes of data at high velocities. Using traditional (semi-) manual methods to gather training data is, therefore, impractical and expensive. Second, humans express more than two emotions, therefore, the typical binary good/bad or positive/negative classifiers are no longer sufficient to address the complex needs of the social marketing domain. This paper introduces a hugely scalable approach to gathering training data by using emojis as proxy for user sentiments. This paper also introduces a systematic Word2Vec based clustering method to generate emoji clusters that arguably represent different human emotions (multi-sentiment). Finally, this paper also introduces a threshold-based formulation to predicting one or two class labels (multi-label) for a given document. Our scalable multi-sentiment multi-label model produces a cross-validation accuracy of 71.55% (± 0.22%). To compare against other models in the literature, we also trained a binary (positive vs. negative) classifier. It produces a cross-validation accuracy of 84.95% (± 0.17%), which is arguably better than several results reported in literature thus far.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| 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".