README.md

XitXat

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Supported Tasks and Leaderboards
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Curation Rationale
  • Source Data
  • Annotations
  • Personal and Sensitive Information
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Licensing Information
  • Citation Information
  • Contributions

Dataset Description

• Point of Contact: Carme Armentano-Oller and Carlos Rodríguez-Penagos

Dataset Summary

XitXat is a conversacional dataset made of 950 chatbot conversations in 10 different domains. Conversations have been commissioned following the Wizard-of-Oz methods. User interactions have been labeled with a intent and rellevant slots. XitXat dataset is released under CC-by licence.

Supported Tasks and Leaderboards

End-to-end dialog systems training, intent classification, anonymisation.

Languages

The dataset is in Catalan (ca-CA)

Dataset Structure

Data Instances

One json file. 10 domains. 950 conversations. 7365 labeled sentences.

Data Fields

For the conversations

• id (str): Unique ID assigned to the conversation.
• nom (str): Name of the conversation.
• número (int): Number of the conversation.
• frases (list): List of the sentences.

For the sentences

• id (str): Unique ID assigned to the sentence.
• text (str): Text of the sentence.
• actor (str): Role. It can ve "agent" or "usuari" (user).
• intent (str): Assigned intent of the sentence.
• slots(list): Slots detected in the sentence.

For the slots

• tag (str): Label of the slot.
• text (str): Text of the slot.
• Start_char (int): Initial character of the span.
• End_char (int): Last character of the span.

Data Splits

• XitXat.json

Dataset Creation

Curation Rationale

We created this dataset to contribute to the development of language models in Catalan, a low-resource language.

Source Data

Conversations were created for this dataset.

Initial Data Collection and Normalization

We commissioned these conversations to a specialized company. The same company did the first annotation. After a new version of the guidelines, we commissioned a revision of the annotations to an independent computational linguist.

Who are the source language producers?

Conversations were created for this dataset following the Wizard-of-Oz method.

Annotations

Annotation process

Annotations where commissioned to a specialized company. They were made following the attached guidelines for intents and slots.

Who are the annotators?

Annotations were commissioned to specialized professional annotators.

Personal and Sensitive Information

No personal or sensitive information included.

There is some feigned personal information, created to give verisimilitude to the dataset.

Considerations for Using the Data

Social Impact of Dataset

This dataset contributes to the development of language models in Catalan, a low-resource language.

Discussion of Biases

[N/A]

Other Known Limitations

[N/A]

Additional Information

Dataset Curators

Carlos Rodríguez-Penagos (carlos.rodriguez1@bsc.es) and Carme Armentano-Oller (carme.armentano@bsc.es) from BSC-CNS.

This work was funded by the Departament de la Vicepresidència i de Polítiques Digitals i Territori de la Generalitat de Catalunya within the framework of Projecte AINA.

Licensing Information

This work is licensed under a Attribution 4.0 International License.

Citation Information

[N/A]

[DOI]N/A