Dark Mode
Video Game Chatbot Dialogue Set
Product Reviews & Feedback
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This resource is an English data-to-text generation dataset created specifically for the video game domain. It focuses on facilitating the development of open-domain chatbots capable of generating conversational responses about various aspects of video games. The data includes structured meaning representations (MRs) alongside conversational target responses, which may express opinions, descriptions, or requests for preferences related to gaming. Its relatively small size, approximately 5,000 examples, makes it suitable for evaluating transfer learning capabilities, low-resource scenarios, or few-shot applications with neural models.
Columns
The dataset is organized in a tabular format and includes two primary content columns.
- meaning_representation: This column contains structured meaning representations (MRs). These MRs capture various aspects related to video games, such as characters, items, locations, and quests. This serves as the input context for model training.
- target: This column contains the conversational text outputs. These target responses are designed to be conversational in nature, offering opinions, descriptions, requests for preferences, or inquiries about game preferences.
Distribution
The dataset is relatively small, consisting of around 5,000 examples or data points. It is typically distributed across different files, such as training and validation splits (
train.csv, validation.csv). Each row represents a distinct example or conversational interaction. The data is clean and highly suitable for model training, although specific row counts for individual files are not explicitly detailed.Usage
This data is ideal for building conversational data-to-text generation models, particularly for open-domain chatbots operating within the gaming space. It can be utilised for training models to engage in natural conversations about video games. The structured meaning representations also serve as valuable inputs for training and evaluating Natural Language Understanding (NLU) models. Furthermore, the conversational target responses, including expressed preferences, can be used to improve personalized game recommendation systems for gamers.
Coverage
The material is exclusively in the English language and focuses entirely on the domain of video games. The structured representations cover a wide range of game facets, providing broad conceptual coverage for game-related conversation. The scope is primarily topical, and there is no specific geographic or temporal time range specified for the collection of the data.
License
CC0 1.0 Universal (CC0 1.0) - Public Domain Dedication
Who Can Use It
- AI Developers and Engineers: For creating open-domain conversational AI systems and chatbots focused on gaming content.
- Machine Learning Researchers: To test model performance in limited data environments, low-resource settings, or few-shot learning tasks.
- Gaming Industry Analysts: To develop better recommendation engines by analysing how user preferences are expressed conversationally.
Dataset Name Suggestions
- ViGGO Conversational Data
- Video Game Chatbot Dialogue Set
- Gaming Data-to-Text Resource
- MR-to-Chat Generation Data
Attributes
Original Data Source: Video Game Chatbot Dialogue Set
QUALITYLoading...