Textile Dataset for Applying into Machine Learning Models (doi:10.7910/DVN/HIZXSS)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Document Description

Citation

Title:

Textile Dataset for Applying into Machine Learning Models

Identification Number:

doi:10.7910/DVN/HIZXSS

Distributor:

Harvard Dataverse

Date of Distribution:

2021-04-19

Version:

1

Bibliographic Citation:

Ngo, Vuong M., 2021, "Textile Dataset for Applying into Machine Learning Models", https://doi.org/10.7910/DVN/HIZXSS, Harvard Dataverse, V1

Study Description

Citation

Title:

Textile Dataset for Applying into Machine Learning Models

Identification Number:

doi:10.7910/DVN/HIZXSS

Authoring Entity:

Ngo, Vuong M. (Trinity College Dublin)

Distributor:

Harvard Dataverse

Access Authority:

Ngo, Vuong M.

Depositor:

Ngo, Vuong M.

Date of Deposit:

2021-04-19

Study Scope

Keywords:

Computer and Information Science, Textile structure, classification

Abstract:

With the help of domain experts we created a data set containing a total of 1,600 fabrics divided into sixteen different categories, each containing 100 items of a particular type of textile. On average, each textile consists of 20,916 vertices, 5,229 hyperedges, 152 terminal nodes, 5,229 regular edges and 10,534 connected edges. The attached file is hypergraph representations which are extracted to k-neighbourhoods and then transferred to vector representations with normal frequence and TF-IDF. These vectors are applied into ML models.

Methodology and Processing

Sources Statement

Data Access

Notes:

CC0 Waiver

Other Study Description Materials

Other Study-Related Materials

Label:

_sawutext_files.zip

Text:

Hypegraph representation of textiles (aslo being SAWU text representation) designed by SAWU Editor in "Gyory. Generic Textile Structure Editor. IGI Global, 2014")

Notes:

application/zip