Manifest

Skidetica technology is a new statistical approach that aims to measure emotions. We believe that sentiments represent discrete values on a probability distribution and that every sentiment can be predicted without historical data.

Most machine learning technologies developed over the past years rely solely on historical data for user intent prediction. Nearly all algorithms used in the modern tech industry require historical data to explain the present or to claim to predict the future. To obtain this data, companies either pay for it or spy on their users.

Skidetica technology offers an alternative. We believe that historical data is not required to predict user intent. There is no need for extensive data collection to predict human emotions; it is sufficient to know the current average pattern of behavior.

Skidetica technology is in active development and requires constant fine-tuning. Given the disproportionate influence of collective human sentiment on stock markets, we decided to use financial markets as a proxy for our human emotion research and to test Skidetica technology.

Skidetica’s Five Theses

• Users’ subjective opinions can be quantified as a continuous probability distribution.
• Users’ moods are discrete values within this distribution.
• In complex environments, historical data is inherently biased and should be used with caution.
• Truth has low resolution and cannot be represented by a single discrete value; it can only be represented as an interval.
• A user-oriented statistical model must be deductive in nature.