Annotate news events with links to previous articles

We are a university research lab working on improving how news is presented to readers. We are developing a system that identifies references to previous events, and automatically finds the article that reported that event in the newspaper's archive. However, the system needs to be trained on human linking decisions. This is where you come in.

Your task will be to identify references to past events in a sample of newspaper articles, and then use an in-built search engine to find the earliest article that reports on that event. For instance, here is a paragraph from the paper we are annotating:

"Mr Dutton walked away from the marginal seat to seek preselection for a safe seat on the Gold Coast after a redistribution by the Australian Electoral Commission rendered his electorate notionally Labor."

There are three events here: Mr. Dutton walking away from his seat, Mr. Dutton seeking preselection for a safe seat on the Gold Coast, and the Australian Electoral Commission's redistribution of his electorate. You would use a search engine to find articles from the same newspaper that report on these events.

Annotating a full article usually involves searches for 5 to 10 events, looking carefully for the best article to link to. You will use an annotation tool that we have built to make the task easier, and follow guidelines we have written on what kinds of events should be linked.

This task requires strong search skills and native quality English language skills. Applicants should have experience reading news articles, and a familiarity with Australian news is a big plus.

Previous annotators have completed about 4 documents per hour, although speed at the task can vary, and about an hour will need to be spent training if you are new to the task. Please bid your fee and time for 25 documents, and we may offer you to annotate more in the future.

We will be hiring multiple people for this project, who will work independent of each other. We have over 200 hours of annotation work to complete in the next month, with further work to come. We hope to build relationships with reliable annotators for future work.

Data Entry Web Search

Project ID: #1232294

About the project