README.md

Relation_Extraction
===================
This repository provides code and additional materials of the paper: "Extraction of the Relations between Significant Pharmacological Entities in Russian-Language Reviews of Internet Users on Medications".

In this work, we trained a model to recognize 4 types of relationships between entities in drug review texts: ADR–Drugname, Drugname–Diseasename, Drugname–SourceInfoDrug, Diseasename–Indication. The input of the model is a review text and a pair of entities, between which it is required to determine the fact of a relationship and one of the 4 types of relationship, listed above.

Data
----
Proposed model is trained on a subset of 908 reviews of the [Russian Drug Review Corpus (RDRS)](https://arxiv.org/pdf/2105.00059.pdf). The subset contains the markup of the following types of entities: and contains pairs of entities marked with the 4 listed types of relationships:
- ADR-Drugname — the relationship between the drug and its side effects
- Drugname-SourceInfodrug — the relationship between the medication and the187source of information about it (e.g., “was advised at the pharmacy”, “the -Drugname-- - Drugname-Diseasename — the relationship between the drug and the disease
- Diseasename-Indication — the connection between the illness and its symptoms (e.g., “cough”, “fever 39 degrees”)
Also, this subset contains pairs of the same entity types between which there is no relationship: for example, a drug and an unrelated side effect that appeared after taking another drug; in other words, this side effect is related to another drug.

Model
----
Proposed model is based on the  [XLM-RoBERTA-large](https://arxiv.org/abs/1911.02116) topology. After the additional training as a langauge model on corpus of unmarked drug reviews, this model was trained as a classification model on 80% of the texts from subset of the corps described above. This model showed the best accuracy on one of the folds of the cross-validation. For additional details see original paper.

How to use
----
See section "How to use" in [our git repository for the model](https://github.com/sag111/Relation_Extraction)

Results
----
Here are the accuracy, estimated by the f1 score metric for the recognition of relationships on the best fold.

| ADR–Drugname  | Drugname–Diseasename | Drugname–SourceInfoDrug | Diseasename–Indication |
| ------------- | -------------------- | ----------------------- | ---------------------- |
| 0.955         | 0.892                | 0.922                   | 0.891                  |