Democratizing the Development of Chatbots to Improve Public Health: Feasibility Study of COVID-19 Misinformation
Bibliographic record
Abstract
BACKGROUND: Chatbots enable users to have humanlike conversations on various topics and can vary widely in complexity and functionality. An area of research priority in chatbots is democratizing chatbots to all, removing barriers to entry, such as financial ones, to help make chatbots a possibility for the wider global population to improve access to information, help reduce the digital divide between nations, and improve areas of public good (eg, health communication). Chatbots in this space may help create the potential for improved health outcomes, potentially alleviating some of the burdens on health care providers and systems to be the sole voices of outreach to public health. OBJECTIVE: This study explored the feasibility of developing a chatbot using approaches that are accessible in low- and middle-resource settings, such as using technology that is low cost, can be developed by nonprogrammers, and can be deployed over social media platforms to reach the broadest-possible audience without the need for a specialized technical team. METHODS: This study is presented in 2 parts. First, we detailed the design and development of a chatbot, VWise, including the resources used and development considerations for the conversational model. Next, we conducted a case study of 33 participants who engaged in a pilot with our chatbot. We explored the following 3 research questions: (1) Is it feasible to develop and implement a chatbot addressing a public health issue with only minimal resources? (2) What is the participants' experience with using the chatbot? (3) What kinds of measures of engagement are observed from using the chatbot? RESULTS: A high level of engagement with the chatbot was demonstrated by the large number of participants who stayed with the conversation to its natural end (n=17, 52%), requested to see the free online resource, selected to view all information about a given concern, and returned to have a dialogue about a second concern (n=12, 36%). CONCLUSIONS: This study explored the feasibility of and the design and development considerations for a chatbot, VWise. Our early findings from this initial pilot suggest that developing a functioning and low-cost chatbot is feasible, even in low-resource environments. Our results show that low-resource environments can enter the health communication chatbot space using readily available human and technical resources. However, despite these early indicators, many limitations exist in this study and further work with a larger sample size and greater diversity of participants is needed. This study represents early work on a chatbot in its virtual infancy. We hope this study will help provide those who feel chatbot access may be out of reach with a useful guide to enter this space, enabling more democratized access to chatbots for all.
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 | 0.044 | 0.082 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.007 | 0.003 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
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".