HomeMiscellaneousStep By Step Beginner Tutorial

Step By Step Beginner Tutorial

In this tutorial, we will walk through the steps of building your first machine learning model with Hopdata. We will build a sentiment classification engine which allows the user to input text and your machine learning model will determine whether the text has a positive or negative sentiment.

Step 1. Download Sample Data Set

For the purpose of our tutorial we will use existing data. The file that we will use is called Twitter Sentiment. The file has two columns. Column A is our categorization column, it labels whether column B is positive or negative. When using HopData, column A always needs to be the column that we are predicting.

To get the data go to our set of example datasets:

example datasets

On the example datasets page, choose twitter-sentiment. To save it, right click on twitter sentiment and save the file on your computer where you can find it. For the purpose of this tutorial, you can also download the twitter sentiment file here .

twitter sentiment


Step 2. Submit Data

Now that you have downloaded a sample data set we are ready to get started. Go to http://app.hopdata.com/create_source.php and select “New Model”

new model

Here you will enter the details of your model and upload the twitter-sentiment file that you just downloaded. Under “Does the file have a header” select “No”. Submit your data to begin the model creation process.

first machine learning model

Step 3: Wait for the model to process

After you have uploaded your data set, you should see the following screen which means that your data is processing.

machine learning processing


Step 4. Evaluate Results

Once your data is finished processing, you will see the following screen. If you look in the model section, you can see that our model is 71% accurate.
model view


Step 5. Test Your New Model  (Validate Results)

If you scroll to the bottom of your individual model you will see that you have the ability to test it out. The sentiment model that we created has the ability to predict whether the text you input is positive or negative and the related weighting of the attributes.

test machine learning model

Congratulations, you have made your first machine learning model! Wasn’t that fun!

Was this article helpful to you? Yes No