Turk Guru offers various modes that alter the filters used after the minimum reward filter, blocked requester's filter and the HIT group filter, filter out newly found HITs. Details of each of the mode are provided below.
Allow all mode: This mode doesn't filter out any HITs. This acceptance mode is useful for high minimum HIT rewards or when the user has a large blocked requesters list. To use it at a lower minimum reward consider reducing the maximum HIT queue limit to 6 or 7 from the Turk Guru console's settings tab under optimization settings. This mode is the fastest as no filtration is done, thereby saving time to send catch requests.
Blocked keywords mode: This mode filters out tasks that contain blocked keywords. If you have some particular type of jobs that you never work on, e.g., transcription tasks, you can add keywords like 'transcribe,' 'transcription' to your blocked list. If you add keywords that are too common in sentences like 'a', 'the', etc., many tasks may get filtered out decreasing the number of HITs that get caught.
Favorite requesters mode: This acceptance mode helps you accept tasks only from your favorite requester's list. Jobs from all other requesters are filtered out. This mode can be helpful If you wish to work on tasks only from specific employers.
Favorite keywords mode: This mode catches tasks that contain a favorite keyword in the title or description. All other jobs are filtered out and not accepted. You can add keywords like "survey", "study", "opinion", "experiment", "academic" to catch survey-like HITs.
Selected HIT types mode: This acceptance mode catches tasks If the HIT is from one of the types selected in the HIT types list. This mode helps workers accept tasks based on easily understandable HIT types giving them more control over their work. This mode's accuracy isn't 100% as it uses Machine learning and some other issues that we face while we build our model.
Past Liked HITs mode: This acceptance mode uses your HIT submissions and queues provided by you like reacting on HITs to accept tasks using Machine learning. This mode improves over time.