AI-ordlista
Den kompletta ordlistan över AI
Anomaly Detection in Streams
Algorithm identifying in real-time observations that deviate significantly from expected patterns in a temporal data stream.
Concept Drift
Gradual or abrupt change in the underlying distribution of temporal data, requiring dynamic adaptation of predictive models.
Temporal Aggregation
Process of reducing temporal granularity by combining multiple data points according to statistical functions (mean, sum, max).
Kalman Filtering
Recursive algorithm for optimal estimation of the state of a linear dynamic system from noisy real-time observations.
Partial Autocorrelation
Measure of correlation between an observation and its past values while eliminating the effect of intermediate observations.
Temporal Resampling
Process of modifying the sampling frequency of a time series through interpolation or aggregation to standardize data.
ARIMA Models in Streams
Adaptive extension of ARIMA models handling continuous time series with dynamic updating of autoregressive parameters.
Wavelet Transformation
Time-frequency decomposition analyzing time series at different scales to capture local and transient patterns.
Temporal Synchronization
Temporal alignment of multiple heterogeneous data streams to ensure consistency in real-time multivariate analyses.
Temporal Vector Quantization
Compression of temporal sequences into optimized discrete representations preserving essential characteristics of the flow.
Temporal Quantile Regression
Conditional modeling of time series quantiles to capture uncertainty and extremes in forecasts.