This is a walkthrough that will show you how easy it is to build a custom extractor and we will give you an example data set to use for this.
In only a couple minutes, you'll be able to see how a machine learns to recognize text right before your eyes. 🤯
We are going to build an extractor that can identify different features (Brand, CPU, RAM, Storage, etc.) in a list of laptop product descriptions.
The end result will be able to extract those features as you can see below. The results are shown to the right, with each tag having its corresponding value – brand: Apple, cpu: Intel Core i7, memory: 8 GB, etc.
Making a laptop feature extractor in five steps
1) Download the list of laptop descriptions here below, it's a CSV file with the first column containing the descriptions. Each description is on its own row.
2) Create a custom model in MonkeyLearn and choose "Extractor".
3) Upload the CSV file that you previously downloaded, and select the column that has the text we want (the first one).
4) Create the following 4 Tags
5) Tag the words that appear in the laptop descriptions by selecting the tags on the right and clicking on the text that they represent (as shown below).
After tagging a couple descriptions, you will see that the model learns to recognize features in the laptop descriptions automatically. It's machine learning at work.
Once you are done tagging, you can test it to see how it works. You can make your own custom extractor just as easily by following the same five steps (if you are on a free account, you can delete the one you just made in your model settings).
For more on this, see our guide on making custom extractors.