Useright: a framework for data quality management in crowdsourcing systems
Bibliographic record
Abstract
There are many problems which cannot be solved by machines, we still need help from people who are intelligent and have skills to solve those problems.Crowdsourcing is the field which uses the skills of people to work on complex problems.In the past decade, many crowdsourcing systems were developed and it is still a growing field among companies and researchers.Though there are many advantages to crowdsourcing, they still lack proper data quality measures.We enumerate the factors that can lead to different data quality problems in crowdsourcing systems.We present a general model for crowdsourcing and talk about different crowdsourcing systems using that general model.This thesis also discusses the classification of crowdsourcing systems using different data quality perspectives.To tackle the data quality problems, we describe the Useright Framework that provides a certification scheme for words and phrases in the document.In this framework, an author publishes documents using credentials obtained from one or more Useright granting institutes.An author asks an institute to provide Useright permissions for a collection of words and phrases that author has chosen.To keep the author's identity private and to check anonymity of author while reading, we introduce the cheap verification scheme.The Useright score for institute is used to increase the believability of document.We provide a design for an extended Useright Framework where an author can edit another author's document using his/her own Useright words and phrases.To evaluate the Useright Framework, we carried out a survey study on Amazon Mechanical Turk, where users were asked to estimate the data quality factors for different documents that were presented in different ways. List ofTables ix 5.2 Variance of Author Qualification for All Documents with All 4 Versions from AMT Workers who spent more than 120 Seconds . . . . . . . . .75 5.3 Variance of Free of Error (Accuracy) for All Documents with All 4Versions from AMT Workers who spent more than 120 Seconds . . . .75
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 | 0.005 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.005 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".