Model Training & Benchmarking
Training
To build our forecasting models, we use publicly available metadata like citation counts, publication dates, and titles.
We primarily source this from aggregators like OpenAlex.
We do not use your data to improve our models.
When it comes to article abstracts, we only train on content with commercially permissible licenses (like CC-BY) or those in the public domain. This data is essential for our hyperparameter tuning and model selection.
Evaluation & Benchmarking [coming soon]
We want our users to trust the numbers. Once a model is trained, we benchmark it against a massive dataset to provide accurate performance statistics. Currently, this includes over 4 million English articles published from 2020 onwards.
We provide benchmarks across as many academic fields and articles as possible. If there are any specific datasets/fields you would like to see that we have not covered, please send a suggestion using the contact form.
Your Personal Data
At NeuroGenesis, your data stays yours.
As a UK-based company, we are fully committed to GDPR and UK Data Protection standards. Your privacy isn't just a legal requirement for us; it's a core value.
CiteScout is built to be local-first. Your data lives on your machine, meaning you can view it even when you're offline.
Note: We are still working on making CiteScout a PWA for full offline access. Currently, if you refresh while offline the app interface will be unavailable until you reconnect.
Browsers can occasionally clear local storage without warning. To guarantee persistence and allow across-device access, we sync your data to our secure cloud.
Due to the costs this incurs on our back-end, this is a feature only available to CiteScout subscribers.
Want to delete your data? You have the right to be forgotten. If you'd like us to delete your account and all associated cloud data, just send a request to privacy@ngenesis.co.uk.
Rights & Licenses
You (or your institution) retain all commercial and publication rights to the data you enter into CiteScout, along with any associated IP claims. We only retain a license to your data for as long as it's needed to provide our services to you.
For the legal specifics, you can always check our Privacy Policy and Terms of Service.