More information:Sentikator Introduction
The Sentikator is an innovative tool that analyzes and quantifies emotions in all types of texts. It provides valuable insights into investment dynamics, forecasts market trends and allows operationalizing behavioral finance for the first time. The technology grew from the fact that the direction of prices is mainly determined by the investors’ mood that it was impossible to coherently measure that. The Sentikator aims to harvest moods and opinions 24/7 and to generate comprehensible and fast sentiment data to be used in financial applications.
The Sentikator is based on the fact that people transport their state of emotion by the choice of words they apply in written or spoken language. Psychological studies were able to proof a strong correlation between physical and mental conditions and the choice of words applied. The Sentikator is able to identify and understand these relationships and patterns and to extract the emotions behind words and texts. Take a look at our Introduction to the Sentikator to learn about the basic approach.
The sentikator consists of several modules that satisfy different purposes.
Input API - the input API can handle almost all type of text-based input sources. It provides the Sentikator with data. Even exotic solutions such as transcription from spoken to written language are possible.
Natural language processor - the Natural Language Processor is a form of artificial intelligence. It can not only recognize text but is even able to comprehend text to a certain degree. It serves to distinguish language, understand grammar and identify important key words or issues within a text.
Sentiment engine - the Sentiment Engine is the core of the Sentikator. It can categorize up to 80 different emotions and content categories within texts.
Analysis of topicality - nothing is older than yesterday's news. The Sentikator identifies whether texts are really new and unique. Trends are detected early and especially event-based traders profit from this feature.
Analysis of relevance - how relevant are incoming news-flows or upcoming events for the prevailing market situation or for current trends? The Sentikator calculates relevant measures of how well texts fit to the topic of interest so that you only get the information you really need.
Frequency analysis - risk and reward in financial markets are closely correlated with the intensity of media coverage. The frequency of how often news sources refer to listed stocks contains comprehensive opportunities for in-depth analysis. The Sentikator measures frequency along several different dimensions.
Output API - you tell us how you need our data and we make it possible. Currently binary data as well as XML, RSS and Json feeds are available by default and even CSV files or formatted images can be provided.
For a comprehensive overview of the technical features please have a look at the Sentikator Scope of Features.