WinterMilestone 4 - Team SneakyLittleHobbitses - Task Ranking

From crowdresearch
Revision as of 19:21, 7 February 2016 by Natasha (Talk | contribs) (Task Ranking : Applying "Filters" to the Task Feed)

Jump to: navigation, search

Team : Sneaky Little Hobbitses

Team members : @natashahervatta, @rajashri92

Task Ranking : Applying "Filters" to the Task Feed

What is the problem?

Crowdsourcing marketplaces like Mechanical Turk and Upwork have proven to be a worthy employment opportunity for workers across the globe. However, there are still a number of user experience issues that they fail to address. One of these issues is that workers need a more efficient way to find the tasks they are looking for. They need to be able to know whether the tasks they’re selecting are worth their time. To solve this issue, Daemo introduces Boomerang, a reputation system that incentivizes alignment between opinion and ratings by increasing the likelihood that the rater will work in the future with any users they rate highly. Thus workers are given easy(?) access to the tasks they would like to work on. In this paper, we would like to explore the possibility of going one step further. What can we do to make it even easier for workers to find relevant tasks at the current time? This is the question that we will try to answer in this paper.

What have other people done?

Search filtering or content filtering is the most natural way of navigating online information space. E-commerce sites are the most common use case for using search filtering, since users are more often than not looking for specific products. Search filtering is also prevalent in search engines like Google and Bing, sites that provide bookings for flights, vacations, deals, and more.

What is the solution?

We propose adding a “Filter” feature to the Task Feed, that works on top of Boomerang. It filters out what is not relevant to the user and drives them to exactly what they’re looking for. There are two basic ways of selecting values for filters : drill-down and parallel selection. The latter allows users to make parallel selections of multiple filter values - such as Time Duration, Price, Requester Rating, etc. We believe this can solve the user experience issue that workers face when it comes to looking for task.