Tools

Content moderation

Does your web content need a discerning eye? Dolores Labs has created an awesome, easy to use, proprietary content moderation service built on Amazon Mechanical Turk called CrowdSifter. Using the variety of tools and APIs provided by Amazon, we built CrowdSifter's interface to streamline the task assignment and completion process for the Turk “workers,” while at the same time allowing us control over cost, quality and speed. This service allows businesses to take full advantage of the cost and scale of Mechanical Turk, without the difficulties of building a customer interface, or the costs of entering into a complex integration. It easily and seamlessly integrates into your existing systems and provides an excellent alternative to traditional content moderation paradigm of dedicated bodies waiting for content.

Go to CrowdSifter
Market research

We make good research results our priority. Crowdsourcing allows us to glean inexpensive marketing results and then our engineers compile the data to make sure that it is useful and necessary to our clients. Companies benefit by getting the information they need at a much lower cost. A user generated example of this model can be seen at ShoeStat.com. ShoeStat gives retailers, brands and advertisers the opportunity to engage particularly passionate consumers in a conversation surrounding the trends in footwear that they love and hate, while giving specific data as to why. Another fun example is our face judging application, FaceStat. We call it market research for the individual.

Go to ShoeStat
Go to FaceStat
Mechanical Turk advantage

Dolores Labs can help you unlock the power of Amazon's Mechanical Turk making it easy, efficient and scalable. Have a task that can be done by an average person? Want to have thousands of average people do it within a day? Collect data with Dolores Labs. We'll help you break your task into small pieces, distribute it to thousands of anonymous workers, ensure their work is high quality, and make the data useful for your business.

Work examples

Sentiment Detection

O’Reilly Media wanted to find the sentiment of thousands of blog comments on a news event. Dolores Labs labeled every comment as “pro” or “con” in under 24 hours.

Dolores Labs also used Turkers to determine what sentiment message board posts about stocks expressed (buy, hold or sell) so that investors could use the data to gauge the community’s feeling about a particular stock.

View Example
Document Classification

Scribd, the “YouTube for documents,” needed to classify the thousands of documents their users upload daily. Knowing whether a document should go in “Health,” “Culture,” “Humor” or one of twenty other categories is a tricky task for a computer, but a very easy one for a human. By using Dolores Labs, Scribd was able to classify thousands of documents and use the data to build an automated classifier.

View Example
Event Classification

Zvents helps people search for fun local events of interest to different users. But they have trouble using automatic classification to label events so that users could find the kinds of things they wanted to attend. Dolores Labs collected 100,000 hand labeled judgments from Turkers that Zvents then used to train their machine learning event classifier.

View Example
Price Extraction

We extracted prices from popular shopping sites, many of which are impossible to automatically scrape accurately.

View Example
Search Relevance

While most search companies employ a staff of annotators who pour over search result after search result to judge search relevance for a given query, Dolores Labs collects the annotation of thousands of different web-users within a number of days. Determining search relevance this way is not only faster and cheaper, but it also allows you to collect judgments from a wide swath of Internet users.

View Example
Image Retrieval

We collected a set of images for products.

View Example