In this video, learn how to deploy a machine learning model created to obtain greater customer insights with SnapLogic Data Science.
SnapLogic Data Science: Model Deployment
Hi! In this video, I will show how Acme Corp can deploy a machine learning model created to obtain greater customer insights with SnapLogic Data Science.
Acme Corp manufactures consumer appliances like washers, dryers, air-conditioners, etc. The company gains insights by deploying a machine learning model built and trained with customer feedback from consumer review sites. By understanding their customer’s sentiments from customer feedback sites, the team at Acme Corp can improve their products and customer experience and also their service.
There are multiple users at Acme Corp that benefit from SnapLogic Data Science, including: Barry – the Business Analyst, Dave – the Data Engineer, and Sam – the Data Scientist.
There are four steps in the data science lifecycle including, data acquisition, data exploration and preparation, model training and testing, and model deployment. We will assume that the data shown has already been prepared and cleansed using SnapLogic Data Science. As Sam, the Data Scientist, I have already built out the model and gone through the testing phase. I can then execute the model with the pipeline called “yelp_03_api. I need to first create an Ultra task. An Ultra Task lets you create REST APIs from SnapLogic pipelines.
I have already created one called “RunSentiAnalysis”. Let’s open it up and see how the “yelp_03_api” has been configured to run as an Ultra Task.
Let’s switch to a webpage that I have created that includes the HTTP endpoint of the Ultra task that has been put behind a load balancer and shall be used run it as an Ultra Pipeline to predict the customer sentiment.
So let’s put in some sample text and then click on ‘Submit’. The feedback is almost instantaneous, and it comes back with a measure of the sentiment in this field called “polarity”. A value of 1 implies that the sentiment is positive so here a polarity value of 0.8 means that the sentiment was quite positive.
Let’s summarize what just happened. After I entered the text into the browser, the Ultra Pipeline executed the model and delivered the results back to the browser. All of this happened in milliseconds without any written code.
Now that either Sam, the Data Scientist or Dave, the Data Engineer was able to deploy the model and can provide richer insights from the customer sentiment analysis to Barry. With these insights, Barry can take the necessary actions to improve the organization’s product innovation and customer experience programs.
By leveraging SnapLogic Data Science along with the platform’s visual, drag-and-drop, no-code interface, Acme Corp is able to significantly increase Data Engineer and Data Scientist productivity while helping improve the organization’s product innovations.
Thank you for watching this video. For more information, please visit snaplogic.com.