Cannabis Personalization: Curating Personalized Cannabis Experiences Through Machine Learning
Bibliographic record
Abstract
With the recent legalization of cannabis in Canada on October 17, 2018, the opportunity for emerging tech to complement and improve the cannabis experience is vast. The legalization of an industry that has been operating in the dark for decades means ample newfound opportunity for government and corporate-funded collaboration, development and research. A specific area of opportunity for growth within the cannabis sector is through personalization. Personalization is often performed via artificial intelligence—specifically machine learning—to develop a customized experience for users on various platforms. This is usually with the intention of targeted marketing. And while mass data collection serves the user by streamlining content to their assumed preferences, which then often directs them to businesses and products, product-tailoring still has vast potential for growth. Though a medical document for cannabis from a health practitioner may include broadband components to look out for, like “THC” and “CBD”, or even suggest ratios of those cannabinoids there is typically no specification on strain type and best consumption methods. Because the effects that cannabis has on a user varies from individual to individual and is dependent on not only their biometrics, but the various other terpenes and cannabinoids that exist in each strain beyond THC and CBD, cannabis users are missing out on opportunities to make the most of their use. Especially for those interested in cannabis to relieve specific symptoms, testing the vast amount of strains that exist and being able to identify the ideal product would be an arduous task on one’s own. Jibed is an app that would use the aggregation of user data to prescribe the most suitable strain of cannabis for that individual based on their specific conditions and body metrics. As the majority of the target audience (cannabis users in Canada) are already logged on to a multitude of data collecting apps (music, health, social, etc.), there is no shortage of data. The app would consider all the implications of the data, from one’s health to mood deduced from the music they're listening to -- just to name some -- in order to achieve optimal prescriptions.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.003 |
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".