expertai/PharmaER.IT

Dataset Card for Dataset Name

The PharmaER.IT Dataset is an Entity Recognition dataset for Pharmaceutical domain in Italian.

The latest is the vesion 0.2.

Dataset Details

Dataset Description

PharmaER.IT is an Entity Recognition dataset in Pharmaceutical domain for Italian.
It has been created using the leaflets of drugs recognized by the Italian Medicines Agency (AIFA).
The data points considered are:

• farmaco: words representing drugs;
• malattia: words indicating diseases;
• sintomo: words indicating a symptom;
• parte anatomica: words representing anatomical parts of the human body;
• O: used for all the remaining words that do not correspond to any of the previous entities.

It is formed by two supervised collections:

• Gold - It consists of 57 documents organized in Train (37), Validation (10) and Test (10) sets.
  The documents were annotated using a semi-automatic procedure with a final HUMAN VALIDATION check.

• Silver - It will consists of a large set of documents automatically annotated
  using the best performing ER model trained and tested with the Gold dataset.

Version 0.2 is the latest.

Dataset Info

• Curated by:
  • Leonardo Rigutini, Stefano Ligabue, Simone Martin Marotta, Marta Spagnolli, Vincenzo Masucci - expert.ai
• Shared by:
  • Leonardo Rigutini, Stefano Ligabue, Simone Martin Marotta, Marta Spagnolli, Vincenzo Masucci - expert.ai
• Funded by:
  • MAESTRO - Mitigare le Allucinazioni dei Large Language Models: ESTRazione di informazioni Ottimizzate” a project funded by Provincia Autonoma di Trento with the Lp 6/99 Art. 5:ricerca e sviluppo, PAT/RFS067-05/06/2024-0428372, CUP:C79J23001170001 12;
  • ReSpiRA - REplicabilità, SPIegabilità e Ragionamento”, a project financed by FAIR, Affiliated to spoke no. 2, falling within the PNRR MUR programme, Mission 4, Component 2, Investment 1.3, D.D. No. 341 of 03/15/2022, Project PE0000013, CUP B43D22000900004;
• Language(s) (NLP):
  • Italian
• License:
  • Apache2.0

Dataset Structure

Each set consists of a json array of objects with:

• document_id: the original document id;
• text: the raw text content extracted from the original PDF;
• tokens: an array of strings representing the tokenized text;
• labels: an array of strings representing the annotation assigned to each token. The data points considered are:
  • farmaco: words representing drugs;
  • malattia: words indicating diseases;
  • sintomo: words indicating a symptom;
  • parte anatomica: words representing anatomical parts of the human body;
  • O: used for all the remaining words that do not correspond to any of the previous entities.

Dataset Creation

The Gold dataset was created following a semi-automatic procedure.
After downloading about 8000 leaflets from the AIFA website, a part of them (67) were labeled using a committee made up of expert systems and LLMs.
The generated annotations were reported on the original documents, highlighting the cases of agreement and disagreement between the committee's models.
The set was divided into 5 "BATCH" that were provided to a team of 5 experts with the task of validating the annotations, by accepting or modifying the proposals inserted by the automatic procedure.
Finally, the resulting dataset was splitted into three sets: train (37), validation (10) and test (10) sets.
The following table reports the distribution of the entities in the gold dataset:

Data Point	Train	Validation	Test	Total
farmaco	5796	1799	99	7694
malattia	2962	939	82	3983
sintomo	2409	745	14	3168
parte anatomica	722	223	50	995
Total	11889	3706	245	15840

Dataset quality assessment

In each BATCH, documents shared, in pairs, with other annotators were inserted.
These documents were used to evaluate some agreement indices in order to provide a measure of the consistency of the annotations in the Gold dataset.
The results are reported in the following table:

Data Point	JPA	CPA	Coverage	k-Cohen
farmaco	0.85	0.91	0.91	0.90
malattia	0.98	0.84	0.86	0.83
sintomo	0.74	0.86	0.87	0.84
parte anatomica	0.68	0.84	0.84	0.76
Average	0.81	0.86	0.87	0.83

BibTeX:

Leaderboard

#pos	Models	Precision	Recall	F1	Accuracy
1°	bert-base-multilingual-cased	0.86174	0.91235	0.88632	0.91235
2°	bert-base-italian-uncased	0.83954	0.91608	0.87614	0.91608
3°	bert-base-multilingual-uncased	0.83896	0.91235	0.87368	0.91235
4°	bert-base-italian-cased	0.83382	0.91314	0.87168	0.91314
5°	SVM	0.81	0.638	0.712
6°	Passive Aggressive	0.756	0.644	0.694
7°	CRF	0.76	0.60	0.664

Dataset Card Authors [optional]

Leonardo Rigutini, Stefano Ligabue, Simone Martin Marotta, Marta Spagnolli, Vincenzo Masucci - expert.ai

Dataset Card Contact

Leonardo Rigutini