Text Data Collection/Writing to Train a Machine Learning Model
Budget: $30 – $250 USD
DISCLAIMER: This task involves dealing with profane language and explicit text content, please proceed with this in mind.
OVERVIEW: I am working on building a machine learning/deep learning model to identify THREATFUL chat messages on a communication platform. In order to build this model, I need to curate pieces of text (a single sentence, multiple sentences, or a paragraph) that contain some kind of THREATFUL message. The objective of this task is to help me build this dataset by writing such THREATFUL examples.
EXAMPLES:
1. If he doesn’t deliver the stuff by tomorrow, I am going to ask John to kill him!
2. He said he was going to kill me if I didn't pay.
3. Keep talking sh*t and I'll take bolt cutters to your fingers.
4. She said she’d poison your pet.
I will share a much more detailed document about what constitutes a THREATFUL message once you have decided to work on the task.
REQUIREMENTS/QUALIFICATIONS:
- Competent in written English
- Comfortable with writing profane/explicit language
- Knowledge of American Urban Slang (the language of the examples needs to be close to chat messages between friends)
DELIVERABLE:
Every annotator must give a Google Sheet containing a single column and “1000” rows. Each row in this sheet corresponds to a single example that is THREATful in nature. An example is a piece of text which could be a single sentence, multiple sentences or a paragraph. The length of each example must be between 10-100 words.
COMPENSATION:
You shall be paid $0.2/example so that comes out to be $200 for the 1000 examples.
OVERVIEW: I am working on building a machine learning/deep learning model to identify THREATFUL chat messages on a communication platform. In order to build this model, I need to curate pieces of text (a single sentence, multiple sentences, or a paragraph) that contain some kind of THREATFUL message. The objective of this task is to help me build this dataset by writing such THREATFUL examples.
EXAMPLES:
1. If he doesn’t deliver the stuff by tomorrow, I am going to ask John to kill him!
2. He said he was going to kill me if I didn't pay.
3. Keep talking sh*t and I'll take bolt cutters to your fingers.
4. She said she’d poison your pet.
I will share a much more detailed document about what constitutes a THREATFUL message once you have decided to work on the task.
REQUIREMENTS/QUALIFICATIONS:
- Competent in written English
- Comfortable with writing profane/explicit language
- Knowledge of American Urban Slang (the language of the examples needs to be close to chat messages between friends)
DELIVERABLE:
Every annotator must give a Google Sheet containing a single column and “1000” rows. Each row in this sheet corresponds to a single example that is THREATful in nature. An example is a piece of text which could be a single sentence, multiple sentences or a paragraph. The length of each example must be between 10-100 words.
COMPENSATION:
You shall be paid $0.2/example so that comes out to be $200 for the 1000 examples.