How to Auto-Label Files Using a Machine Learning Model?

Thanks to social media and the internet, we have millions of text data to analyze in these days. Suppose you’re constantly tracking consumer feedbacks about your brand, your competitors and products related to your business. In that case, you should analyze those feedback month to month to understand where your consumers are shifting. 

Reading all the conversations that mention your brand, your competitors and analyzing these conversations manually is impossible. But Machine Learning can do that job for you. 

Let’s dig in to see how you can auto-label and classify consumer feedbacks or any text data by using a machine learning model. 

First, Sign in to your Kimola Cognitive Account.

Sign in to your Kimola Cognitive account to work on your file. You will be directed to cognitive.kimola.com after signing in.

Use Custom ML Models or Choose Pre-Built ML Models:

There are 2 ways at Kimola Cognitive to auto-label your text data: 

a- Use Custom Machine Learning Models

You can train, store and use your custom machine learning models at Kimola Cognitive and auto-label your data with those models. Creating custom machine learning models is the best option if you’re constantly tracking data and want to keep track of your own metrics by analyzing your data daily, weekly or monthly. 

* If you don’t have a custom machine learning model and want to create one, read this article on How to Create a Custom Machine Learning Model at Kimola Cognitive. 

b-  Use Pre-built ML Models from Kimola Cognitive Gallery

Kimola publishes public machine learning models at Kimola Gallery and you can use them with every plan! See Kimola Cognitive Gallery and our ready to use machine learning models here.

Using Custom Machine Learning Models at Kimola Cognitive

After signing in to Kimola Cognitive, find “Upload Your Dataset!” on the home page; it’s a big orange cloud with an upload icon and it’s impossible to miss!

Let’s Start Auto-Labeling! (You can watch our video below!)

• Drag & drop your file (.xls or .csv) that you’d like to be auto-labelled to the “Upload your dataset!” space. After dragging & dropping your file, a big white onboarding screen will welcome you.
• Choose the text column to start auto-labelling and press Continue.
• The next screen will detect the language of your text. (See supported languages here.)
• Now, you will see the Machine Learning Models that you can use. On the left side, you can see your custom models and the Gallery next to it.
• You can choose more than one model at Kimola Cognitive! For example, you can select your own custom ML model from “Models”, and you can also select “Sentiment Classifier Model” from the gallery. 
• Click Continue if you selected your models. 
• Give a title to your report and check the Language and ML Models you chose. 
• Create The Report! 
• Wait for it, aaaand your report is ready! Click to view your report. 

See it in action: