Brand_Description_Data

Introduction

1475 brands with their descriptions from different sources are included in the file.

Standardized Data

Raw Data (Corpus)

Contents

File	Description
./brands.json	Data

Format

All files are in json format. Here is a sample:

[
    {
        "Brand_Name": "Python",
        "Category": "Programming Language",
        "Description": [
            {
                "Source": "Wiki",
                "Content": "Python is an interpreted, high-level, general-purpose programming language."
            },
            {
                "Source": "https://www.python.org/",
                "Content": "Python is a programming language that lets you work quickly and integrate systems more effectively."
            },

            ...
            ...
            ...

        ]
    },

    ...
    ...
    ...

]

Current Usages

The data is currently used to test keywords extraction related programs.

Issues

• The context from Wikipedia may have problems with the format that some sentences are splitted by line break (\n) but not Whitespace Character ( ). In some cases, two words will be regarded as one during word segmentation (if the program first delete all line break and the words immediately before and after the line break are joint).
• Some brands are unfortunate to have no context crawled.