Bitext and UiPath

Bitext has partnered with UiPath to bring you the first multilingual topic-based sentiment analysis tool designed specifically for integration with RPA.

With support for more than 8 languages and designed for easy “no-code” adaptation to any vertical, it is the only tool with an SLA commitment to reaching 90% accuracy in the first 12 months after deployment.


Bitext Topic-based Sentiment Analysis in UiPath AI Center

Bitext’s multilingual topic-based sentiment analysis tool is now available as an ML skill in the UiPath AI Center. This means that you can call it from your flows in UIPath Studio Pro, so you can now easily set up an end-to-end process to mine review data and extract actionable insights – using RPA bots to scrape reviews and our sentiment analysis tool to analyze them.

You can collect customer reviews from multiple sources, extract topics and their corresponding opinions, and then aggregate the topics into categories you define. You can then see not only which categories are most talked about in the reviews, but also which ones have more positive or negative reviews.

Thanks to integration with the Insights Dashboard, you can take advantage of further automation and also set up notifications and alerts to monitor the sentiment for topics and categories.

One Case Study: How to Prioritize Investments

How to decide where to invest your money to improve your business user experience and revenue

  • Define relevant categories for your business
  • Scrape reviews from multiple sources
  • Identify which topics have bad reviews to invest in them
  • Identify which topics have good reviews

Base this process on hard evidence – actual reviews

Customer Insight in minutes. Movistar Telefónica saves 75% with the Bitext Linguistic Engine


Customer and Problem

Movistar, part of Telefónica/O2, an international telecommunications company
End user:
Movistar’s global marketing teams
To gain fast visibility of what customers are saying about Movistar products, services and competitors on social media and the web, so that insights can be used to improve global marketing strategy

Previous Approach:
How did the work get done before Bitext?


  • Marketing team members were tasked with manually reviewing sometimes as many as hundreds of thousands of mentions
  • Objectivity was hard to achieve. Patterns were difficult to detect

Analyze feedback consistently


  • Analyze 8 languages at once (for this project Spanish, English and Portuguese)
  • Gain clear insight into customer feedback in real time
  • Quickly and easily extract main topics and related customer sentiment from massive volumes of unstructured text
  • Determine the intensity of the sentiment expressed

Bitext and UiPath AI Center Approach


How Bitext resolved the challenge:

  • Checking of all languages simultaneously
  • Automated topic-level sentiment used to quickly differentiate between positive, negative and neutral comments. Always providing the sentiment topic
  • Sentiment assigned based on domain-specific ontologies
  • Grading system used to identify intensity of sentiment

Why Bitext technology is the right solution:

  • The Bitext linguistic engine has deep knowledge of 8 languages and is able to understand sentence structure


  • Real-time analysis that gives marketers results in minutes
  • Bitext enables consistency across languages (the same rules apply to all responses) and documented -results (granularity)
  • Bitext is easy to customize
  • Accuracy is objectively measurable
  • Bitext handles all maintenance of its engine
  • UiPath AI Center and UiPath Action Center enable continuous retraining of the sentiment analysis model with human-validated data


    Camino de las Huertas, 20, 28223 Pozuelo
    Madrid, Spain


    541 Jefferson Ave Ste 100, Redwood City
    CA 94063, USA