LocalOutlierFactor

Local Outlier Factor for outlier detection Compatible with sklearn.neighbors.LocalOutlierFactor

Detects outliers using local density deviation. LOF > 1 indicates outlier (lower local density than neighbors).

Example

const lof = new LocalOutlierFactor({ n_neighbors: 20, contamination: 0.1 });
lof.fit(X_train);
const predictions = lof.predict(X_test);  // -1 for outliers, 1 for inliers

Constructors

Constructor

new LocalOutlierFactor(options?): LocalOutlierFactor

Defined in: ds/src/ml/outliers.js:672

Parameters

options?

algorithm

string = "auto"

‘auto’ (only option for now)

contamination

number = 0.1

Expected proportion of outliers (default: 0.1)

metric

Function = null

Distance function (default: euclidean)

n_neighbors

number = 20

Number of neighbors (default: 20)

novelty

string = false

If true, can predict on new data (default: false)

Returns

LocalOutlierFactor

Properties

_tableColumns

_tableColumns: any

Defined in: ds/src/ml/outliers.js:696

algorithm

algorithm: string

Defined in: ds/src/ml/outliers.js:685

contamination

contamination: number

Defined in: ds/src/ml/outliers.js:687

label_column

label_column: any

Defined in: ds/src/ml/outliers.js:689

metric

metric: Function

Defined in: ds/src/ml/outliers.js:686

n_neighbors

n_neighbors: number

Defined in: ds/src/ml/outliers.js:684

negative_outlier_factor_

negative_outlier_factor_: any[]

Defined in: ds/src/ml/outliers.js:692

nFeatures_

nFeatures_: number

Defined in: ds/src/ml/outliers.js:695

novelty

novelty: string

Defined in: ds/src/ml/outliers.js:688

offset_

offset_: number

Defined in: ds/src/ml/outliers.js:693

threshold_

threshold_: number

Defined in: ds/src/ml/outliers.js:694

X_

X_: any[][]

Defined in: ds/src/ml/outliers.js:691

Accessors

negative_outlier_factor

Get Signature

get negative_outlier_factor(): number[]

Defined in: ds/src/ml/outliers.js:917

Get negative outlier factor for each sample

Returns

number[]

Negative outlier factors

Methods

_pairwiseDistances()

_pairwiseDistances(X): any[][]

Defined in: ds/src/ml/outliers.js:840

Compute pairwise distances

Parameters

X

any

Returns

any[][]

fit()

fit(X): LocalOutlierFactor

Defined in: ds/src/ml/outliers.js:704

Fit the model

Parameters

X

any

Training data (2D array, {data, columns}, or array of objects)

Returns

LocalOutlierFactor

this

fit_predict()

fit_predict(X): number[]

Defined in: ds/src/ml/outliers.js:900

Fit and predict in one step

Parameters

X

any

Data

Returns

number[]

Predictions: -1 for outliers, 1 for inliers

fit_transform()

fit_transform(X): any[] | number[]

Defined in: ds/src/ml/outliers.js:909

Fit and transform in one step (parity with IsolationForest).

Parameters

X

any

Data

Returns

any[] | number[]

Labels or table with outlier column

predict()

predict(X): number[]

Defined in: ds/src/ml/outliers.js:860

Predict if samples are outliers

Parameters

X

any

Data (must be training data if novelty=false)

Returns

number[]

Predictions: -1 for outliers, 1 for inliers

transform()

transform(X): any[] | number[]

Defined in: ds/src/ml/outliers.js:888

Transform data by adding outlier labels (parity with IsolationForest). Table or array-of-objects input -> original rows augmented with the label column, realigned to every original row (rows dropped for missing values default to inlier). 2D-array input -> bare -1/1 label array (sklearn-style).

Parameters

X

any

Data

Returns

any[] | number[]

Labels or table with outlier column