Abstract: Techniques for predicting an impact of one or more events on service demand are disclosed. Some embodiments include first and second sets of data characterising properties of historic events using metadata tags, and demand for services that are then filtered to distinguish ordinary demand from extra-ordinary demand. Machine learning is used to determine correlations between metadata tags and extra-ordinary demand to produce a third data set operable for predictive determinations of future event impact on service demand.

Abstract: Techniques for an iterative singular spectrum analysis are provided. In one technique, a first analysis, of time series data, is performed that results in a first reconstructed version of the time series data. The first analysis, of the time series data and a portion of the first reconstructed version, is then performed that results in a second reconstructed version of the time series data. Based on a termination condition, it is determined whether to perform the first analysis relative to a portion of a third reconstructed version of the time series data. A second analysis, of the time series data and a portion of a fourth reconstructed version of the time series data, is performed that results in a fifth reconstructed version of the time series data. The second analysis is different than the first analysis. A difference between the time series data and the fifth reconstructed version data is computed.

Abstract: Techniques for regional viewership predictions of broadcast events such as live broadcast professional sporting events. The techniques can make the predictions without a direct response variable such as regional viewership data for training a prediction model. Instead, in one technique, demand information for a good or service is used. From the demand information, a derivative demand for the good or service relative to a normal demand is determined. A regression framework is used to learn relationships between the derivative demand for the good or service and features of past live broadcast sporting events. This results in a matrix of feature weights. A non-parametric mixture framework is then used to find a set of feature weights that can be applied to features of future broadcast events to generate regional viewership predictions for the events.

Abstract: Techniques for predicting an impact of one or more events on service demand are disclosed. Some embodiments include first and second sets of data characterising properties of historic events using metadata tags, and demand for services that are then filtered to distinguish ordinary demand from extra-ordinary demand. Machine learning is used to determine correlations between metadata tags and extra-ordinary demand to produce a third data set operable for predictive determinations of future event on service demand.