Data · dataset · 2016
AmazonQA
Listed in BERD@NFDI Data Portal
The AmazonQA dataset is a large‑scale, review‑based question‑answering collection built on the Amazon product and QA corpora, containing 923 k questions, 3.6 M answers, and 14 M review snippets drawn from 156 k products.
Description
Each entry is stored in JSON‑L format and includes the question text, question type (yes/no or descriptive), answer texts with helpfulness counts, product identifiers, category labels, and an answerability flag indicating whether the question can be answered from the available reviews.
The repository provides preprocessing pipelines for multiple data configurations, analysis notebooks, and reference implementations of the baseline models described in the accompanying paper.
Links
Where it is published
- BERD@NFDI Data Portal record berd-platform.de/records/m31kq-ed179 ↗
landing page · from berd platform de
Catalogue records · 1
- InvenioRDM API berd-platform.de/api/records/m31kq-ed179 ↗
metadata API · from berd platform de
Topics
- Stated by source
- Machine-learning · Product information · Product testing
- From keywords
- Computer Science & AI · Economics & Finance · Machine learning
- Inferred from text
- Text 75%
Provenance · 1 source records, 12 field assertions
| Source | Key | Last seen | Raw |
|---|---|---|---|
| BERD@NFDI Data Portal | m31kq-ed179 | 12 d ago | JSON v1 |
| Field | Assertion | Extractor | Evidence |
|---|---|---|---|
| access_level | source · berd platform de | connector:berd_platform_de@1.0.0 | |
| concepts[field].anzsrc:group:4611 | mapping · berd platform de | vocabulary-mapper@1.0.0 | keywords['Machine learning'] |
| concepts[field].local:field:computer-science-ai | mapping · berd platform de | connector:berd_platform_de@1.0.0 | |
| concepts[field].local:field:economics-finance | mapping · berd platform de | connector:berd_platform_de@1.0.0 | |
| concepts[modality].local:modality:text | enrichment · berd platform de | keyword-concept-rules@1.0.0 | title+description (75%) |
| concepts[topic].invenio_subject:berd_platform_de:machine-learning | source · berd platform de | connector:berd_platform_de@1.0.0 | |
| concepts[topic].invenio_subject:berd_platform_de:product-information | source · berd platform de | connector:berd_platform_de@1.0.0 | |
| concepts[topic].invenio_subject:berd_platform_de:product-testing | source · berd platform de | connector:berd_platform_de@1.0.0 | |
| description | source · berd platform de | connector:berd_platform_de@1.0.0 | /metadata/description |
| publication_date | source · berd platform de | connector:berd_platform_de@1.0.0 | |
| title | source · berd platform de | connector:berd_platform_de@1.0.0 | /metadata/title |
| updated_date | source · berd platform de | connector:berd_platform_de@1.0.0 |