Privacy Policy

Controller
Alice Laquerrière-Hecker
II. Rote-Haag-Weg 12
52076 Aachen, Germany
E-Mail: alice.laquerriere@gmx.de

1. General Information

EvidenceEngine is a scientific literature search platform designed for academic and research use. The application operates entirely within the European Union and complies with the requirements of the General Data Protection Regulation (GDPR). No user accounts, tracking tools, or marketing cookies are used. The integrated AI model (TinyLlama) runs fully locally on the server and does not transmit, store, or analyze any personal data externally.

2. Principle of Personal Data

You can use this website without disclosing your identity. Personal data is collected only if you voluntarily provide it to us (e.g., by contacting us via email) or if it is technically necessary to use this website. The latter particularly includes the IP address transmitted by your device; however, this data is not used in a personally identifiable manner.

3. Processed Data within EvidenceEngine

4. Local Language Model (LLM)

The integrated language model TinyLlama runs entirely on the local server. No input data, IP addresses, or other user information are transmitted to third parties or external systems. All processing takes place only in volatile memory (RAM) during runtime and is immediately discarded after the response is generated. The model does not log, store, or analyze any user data, nor does it perform remote or external processing.

5. Legal Basis for Processing

6. Data Processing by Third Parties

If external hosting services are used, data processing occurs exclusively within the EU under a Data Processing Agreement (Art. 28 GDPR).

7. Storage Duration

8. Rights of Data Subjects

Under Articles 15–21 GDPR, you have the right to access, rectify, delete, restrict, or object to the processing of your personal data, and to request data portability. Requests can be directed to the contact details provided above.

9. Sources and Licenses

EvidenceEngine exclusively uses open-access sources (OpenAlex, Europe PMC, DOAJ, PubMed Central, arXiv). For each record, the DOI, source, and license are displayed where available. No copyrighted full texts or paywalled materials are stored or mirrored.

10. Security

All data transmissions are encrypted via HTTPS/TLS. The system is secured using modern protective measures including firewalls, rate limiting, and up-to-date security patches to prevent unauthorized access.

Back to search