Hello Turks! We are NY Messaging (formerly Calvin B Kim), Requester ID: A1K8JY6NY1WWTV. We need assistance in identifying short messages which are potentially spam, results of which will be used for a Machine Learning based classifier. The task is very simple; you will be presented with a number of messages that need to be classified as Valid or Spam with an additional option of "I don't know" if the message is tricky to classify. The task should not take more than 2-3 minutes although we do provide plenty of more time, just in case. The name of the task is "Spam SMS Identification" with the payout of $0.20 for 10 easy classification questions per HIT [Number of questions may change periodically with the payout being adjusted in proportion] [We are willing to change the payout per HIT if workers think it is too low for the amount of work. Suggestions are most welcome] . A large number of HITs will be published on an ongoing basis so there will definitely be plenty to go around. There is also a bonus in store for workers who complete a large number of our HITs successfully! You would need to take a Qualification first [messageclassificationnda] before you can access the HITs. Any suggestions to make the HIT more worker friendly is appreciated. Note: Earlier, we made an error in judgement by adopting a 'Known answer' policy which resulted in several assignment rejections. We used this policy as a measure to check the quality of answers provided by workers by choosing questions which we thought were easily classifiable into the 2 groups. However, we failed in this task and sincerely apologize to workers who were affected by it. This oversight generated a lot of negative feedback resulting in poor response towards our HIT. We have tried to learn from these mistakes and hope that the HITs are more worker friendly. The 'known answer' policy is now completely removed. Thank you All!
