Automatic Text Analysis

Natural Language Processing

IIC technologies based on Computational Linguistics and Statistics allow us to offer services based on Automatic Text Analysis. We perform estimations and automatic analysis of any kind of text with the aim of revealing, classifying or finding non-explicit content in order to simplify the task of people who accomplish it manually.

Texts may be produced in different formats (platforms, databases, TXT, Word, PDF, etc.), come from different channels (written documents, webpages, comments on internet forum or social networks, tweets, phone-calls transcripts, email from clients, etc.), consist of only some words or be very long.

What do we do?

We analyse natural language texts automatically in order to:

  • Grasp meaning:
    • Opinion. For instance, positive or negative.
    • Sentiment. For instance, happiness, despair, desire, surprise, etc.
    • Intention. For instance, questions, doubts, complaints, suggestions, buying intentions…
    • Awareness. For instance, whether citizens are involved in social causes or not.
  • Detect the main topic or words of a conversation.
  • Classify in categories.
  • Infer demographic variables such as gender and age.


Computational Linguistics solves many aspects related to the content of a message. Thus, if we analysed call centre conversations, we might:

  • Categorise complaints and incidences: categorise complaints according to the classification defined by the client.
  • Analyse sentiment and emotion, consider…
    • Which are the topics that generate the most negative —or positive— feelings?
    • What is the most prevailing emotion among clients: joy, curiosity or compassion?
    • How do the call centre operators express themselves? Which one does communicate warmly and express emotions?
    • How do the clients express themselves? Are they aggressive?
  • Learn from the best practices when dealing with clients analysing the most successful conversations.
  • Learn more about the client’s profile to redirect him to the most suitable operator.
  • Identify and relate client behaviours to scores from users’ surveys.

Why invest in Natural Language Processing?

Automatic analysis reports obvious advantages as opposed to a manual classification, especially when dealing with large volumes of texts:

  • Substantial reduction of costs associated to time and staff required for a task.
  • Resource planning is improved.
  • Process information in real time as it is a high-performance system for large volumes of texts.
  • Speeds up decision-making processes.
  • It is adjusted and personalised to the client’s needs.

It is possible to identify and relate client behaviours to scores from users’ surveys.