Kernel density estimate
A kernel density estimate is a nonparametric probability-density estimate formed by combining localized kernel functions centered on observed data points.
Outlier Detection in Python
A kernel density estimate is a nonparametric probability-density estimate formed by smoothing contributions from sample observations.
- How it is used
- Uses estimated density to flag sparse numeric values, including internal outliers between groups.
- How deep it goes
- Explains kernels and bandwidth visually, then works through fitting, scoring, and thresholding; does not derive the estimator mathematically.
- Examples
- Uses segment-data hue-mean scores, bandwidth illustrations, and repeated k-means clusterings.
- What you can do after
- Apply it via scikit-learn’s KernelDensity to score one column, or via PyOD’s KDE detector to obtain training scores.
Read how Outlier Detection in Python explains kernel density estimate →
Machine Learning Algorithms in Depth
A nonparametric technique that estimates a probability density by combining localized kernel functions centered on observations.
- How it is used
- Uses localized kernels to estimate probability densities for accurate analysis of relatively small datasets.
- How deep it goes
- Explains Gaussian and general kernels, bandwidth smoothing, trade-offs, and implementation, but defers efficient tree-based evaluation.
- Examples
- Applies Gaussian KDE to RNA-seq data for estimating T7 promoter flux and compares it with a histogram.
- What you can do after
- Apply KDE to estimate and compare data densities, tune bandwidth smoothing, and recognize memory and evaluation costs.
Read how Machine Learning Algorithms in Depth explains kernel density estimate →
Related concepts
- bandwidth parameterconfigured by · A bandwidth parameter is a scale or radius controlling the spatial range of kernel weighting and consequently the locality and smoothness of an estimate.
- kernel functionuses · A kernel function is a mathematical function that measures similarity between pairs of inputs, often representing an inner product in an implicit feature space.
- curse of dimensionalitycontrasts with · The curse of dimensionality is the set of statistical, geometric, and computational difficulties that tend to increase as the number of dimensions grows.
- mahalanobis distancealternative to · Mahalanobis distance is a covariance-adjusted measure of separation between observations or between an observation and a reference distribution.
- synthetic dataproduces · Synthetic data is data generated artificially rather than collected directly from the target process, often to approximate its properties or meet specified requirements.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.