Introduction
Before you start the Brand extraction process, there are a number of settings you can adjust to influence the behaviour of the process. These settings influence the brands that are extracted, the shape of codeframe that is created and the autocoding that is applied.
The new codes are calculated by grouping together typos and spelling mistakes according to the specified settings.
Settings
You can adjust the following options:
Options | Description |
---|---|
Minimum Mentions | Minimum number of distinct verbatims which contain the brand name before a code is created. |
Typo Matching | Choose an option from Exact Match to 3 Letter Difference to limit the number of character differences between 2 brand names to be considered the same. For example, "Mc Do" and "Mac Do" have 1 Letter Difference. Whereas "Mac Donald" and "Moc Dinald" have 2 Letter Difference. |
Brand Separators | (optionnal) Controls which characters separate brand in a multiple mention question. By default, the Brand Extraction process will use ";" , "," and "/" (semi-colon, comma and slash) |