Data In, Insights Out: Analyze Sentiment in Oracle Analytics Cloud

 Author: Philip Godfrey

Oracle Analytics Cloud is a powerful tool that enables organizations to make data-driven decisions by providing advanced analytics capabilities.

Previous blogs in this series include getting started with Oracle Analytics Cloud, loading data, and creating data flows, and our focus this week is on Analyze Sentiment.

In our previous blog post, we created a data flow, and added in Analyze Sentiment step, but I wanted to explain what this is, how it works, and what it can add to your data.

 

What is Sentiment Analysis?

Sentiment Analysis is a subfield of Natural Language Processing (NLP) and Machine Learning that involves analyzing text data to determine the emotional tone or attitude conveyed by the writer or speaker.

The goal of sentiment analysis is to automatically classify text as expressing a positive, negative, or neutral sentiment.

If we consider a review from a restaurant, the review might say something like:

 

"The food was amazing, but the service was terrible! I would absolutely visit again”.

 

Sentiment Analysis in this case will help us figure out that the reviewer is overall happy with the food, and would return, but unhappy with the service.

It becomes complex because human language is complex! If one word was changed in this review, it becomes completely different.

 

"The food was amazing, but the service was terrible! I would absolutely visit again, not!”.

 

At this point, we can see the reviewer is happy with the food, but unhappy with both the service, and would not visit again, although this could be difficult to pick up as it’s a sarcastic comment, and humans often express themselves in ways that are open to interpretation.

 

Analyze Sentiment

Going back to our example in the previous blog, once the Analyze Sentiment step has been added into the Data Flow, we need to indicate the column we want to understand the sentiment of through parameter selection.

In the Disney dataset, this will be Review_Text.




As the screenshot above suggests, this step will create a column, named emotion, and will append this to our Disney Review Dataset.

As we can see here, each review has been reviewed by Oracle Analytics Cloud, and provided an emotion:

  • Positive
  • Negative
  • Neutral


Each review will now be classed as either Positive, Negative, or Neutral based on the language content, semantic constructs, and context.

What are the benefits of understanding sentiment?

Understanding Sentiment Analysis has numerous benefits across various industries and applications.

Here are some of the most significant advantages:

·         Improved Customer Experience: analyzing customer feedback allows businesses to identify areas for improvement, resolve issues promptly, and enhance customer satisfaction.

 

·         Informed Decision-Making: Sentiment Analysis helps organizations make data-driven decisions by providing insights into customer opinions, preferences, and emotions.

 

·         Market Research: inform market research, helping businesses understand consumer opinions and sentiment about products, services, or competitors.

 

·         Brand Reputation Management: By monitoring online conversations, businesses can identify and address negative sentiment quickly, protecting their brand reputation and building trust.

Look out for the next blog in the series which showcases the newly created dataset, which incorporates our Disney Review Data with Analyze Sentiment and we take this into OAC to explore using Augmented Analytics through Explain functionality.


Comments