Using weak supervision to scale the development of machine-learning models for social media-based marketing research
Bibliographic record
Abstract
Marketers have expressed substantial enthusiasm about the potential of social media data to enhance marketing research, and the computer science literature provides many examples of using the text and network connections of social media users to infer measurements of interest to marketers. Despite this, the adoption of such machine-learning approaches has been surprisingly limited in marketing practice, in part due to the hurdle of procuring the labelled training data typically used to build such models. This paper discusses how the organic structure of social media can often be leveraged to circumvent the need for such curated data labels. It describes two emerging methodological themes of weak supervision — training on exemplars and training on groups — that are broadly promising towards this goal, providing examples of how they have been applied towards a variety of marketing tasks without requiring any manually labelled training data, and in some cases, requiring nothing more than a single keyword as input. This paper presents these approaches in the hope that examples will inspire and facilitate the development of a broader range of flexible, scalable and cost-effective models for social media-based marketing research, and stimulate additional research in this area.
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 | 0.018 | 0.057 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.007 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.002 | 0.007 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".