Scoring / regression models
Scoring models (also known as regressors) are used to predict a continuous value based on input data. They are used in a variety of applications, such as predicting the price of a house based on its features, predicting the temperature based on weather data, or predicting the number of sales based on historical data.
The TrueState platform supports the following scoring models:
- Text-based preference models: These models mimic a human's preference by learning to discern between favourable and unfavourable records. For example, a model that scores investment proposals based of manually labelled preferences from investment professionals. These models are interestingly used in the traning of conversational text models to ensure more natural sounding outputs (in this context they are referred to as a reward model).
- Tabular scoring models: These models predict a continuous value based on structured data records (e.g. rows in an excel table). For example, a model that predicts the price of a house based on its features (e.g. location, size, number of rooms).
Use-cases for scoring models
Scoring models are used in a wide range of applications, including:
Price prediction: Predicting the price of a product or service based on historical context data. For example, predicting the price of a house based on its features (e.g. location, size, number of rooms). Demand forecasting: Predicting the demand for a product or service based on historical sales data. For example, predicting the number of sales for a product based on historical sales data.