Common problem:

Analyzing data to extract insights is usually done manually

Everyday businesses receive from its customers plenty of opinions, experiences, complaints or feedback via very different platforms. Collecting them from multiple sources is not a problem, but analyzing the data to extract the insights can be quite arduous and time-consuming if you do it manually.


Sentiment Analysis is the tool to use to discover what your customers are saying about you

Bitext Topic-based Sentiment Analysis service provides polarity, sentiment scoring, sentiment text and, most important, sentiment topic identification out of raw data with over 90% accuracy by relying on Deep Linguistic Analysis and in our in-house parsing technology.

Sentiment Analysis

I want to try it

Use Case: Twitter Sentiment Analysis

People talk and they like to express their opinions in public. When it comes to a marketer this is a great source to get data and turn it into insights. However, analyzing many comments from Twitter, for example, can be time-consuming: analyzing the phrases, categorizing them depending on the topic, what do they say etc.

If you want to know how easy is to solve this task with Bitext Sentiment Analysis, download our use case.

Download a Real Case


Extracts as many opinions as the sentence contains

The sentiment analysis service is not limited to extract a single opinion per sentence. It can actually detect as many opinions as the sentence contains. As an example:

“This phone is awesome, but it was much too expensive and the screen is not big enough”

Three opinions will be extracted:

  • “phone” + “awesome”

  • “phone” + “much too expensive”

  • “screen” + “not big enough”

Accurately handles complex language structures to extract the polarity

Complex language structures, such as negation or comparative sentences, play a major role in sentiment analysis. Deep Linguistic Analysis automatically handles these structures and can capture the difference between opinions like:

  • “This phone is much better than my old phone.” - Positive

  • “This phone is not much better than my old phone.” - Negative

Identify accurately the topic of the sentiment

We don’t focus on just telling you if something is positive or negative. We go beyond that and show you what the opinion is about - the sentiment topic. This is really useful to quickly find problem areas in your business and to get really granular results without reading the original texts.

For example, “The service was horrible but the food was absolutely brilliant.” We have two different opinions about two really different points. The sentiment topics are: “service” and “food”.

Capturing the intensity in sentiment scoring

Sentiment scoring is also based on Deep Linguistic Analysis. The more intense the feelings of the person about the subject, the higher or lower the score.

To achieve this, the linguistic analysis detects linguistic features such as the semantic strength of the vocabulary or the use of intensifiers like “really”, “very” or “extremely”. For example, “Installing software on this machine is painful!” will be scored as less negative than “Installing software on this machine is really very painful indeed!”.

If you want additional info schedule your demo

bitext madrid offices


José Echegaray 8, building 3, office 4
Parque Empresarial Las Rozas
28232 Las Rozas

san francisco bitext offices


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