Timeline for Attitude to text mining and preparing tokens, irrelevant words, low accuracy
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 23, 2019 at 14:55 | comment | added | Erwan | Imho the word "dictionary" is fine but it can make people think that there is a definition or something associated with each word. Technically if only words from this list are taken into account then it can be called the "vocabulary", but it's also common to just call this a "list of terms" (with variants: list of target/domain terms, etc.). | |
| Nov 23, 2019 at 14:31 | comment | added | heisenberg7584 | and does it have any "professional" name ? Is it called dictionary like I use the term or maybe somehow different? | |
| Nov 23, 2019 at 14:15 | comment | added | Erwan | Sure it's ok, the use of terminology resources is common in many specific domains. The only problem with that is that it often requires a lot of work to create it and maintain it. | |
| Nov 23, 2019 at 14:00 | comment | added | heisenberg7584 | and is it OK to prepare a dictionary of words that are relevant to problem and then before classification keep just words from dictionary? Or we don't do this in text-mining and it is not a good approach? | |
| Nov 23, 2019 at 13:49 | history | answered | Erwan | CC BY-SA 4.0 |