Wolfram Language Paclet Repository
Community-contributed installable additions to the Wolfram Language
AntonAntonov`MonadicGeometricNearestNeighbors`
GNNMonFindAnomalies |
| GNNMonFindAnomalies[ pnts pnts | | ||||
|
| | ||||
Details and Options
Examples
(2)
Basic Examples
(1)
In[1]:=
SeedRandom[454];dsPoints=ResourceFunction["RandomTabularDataset"][{30,{"X","Y"}},"Generators"1NormalDistribution[0,5],2NormalDistribution[12,3],"RowKeys"True];dsPoints〚1;;4〛
Out[1]=
Here is a monadic Geometric Nearest Neighbors Monad (GNNMon) object is created:
In[2]:=
gnnObj=
[dsPoints]⟹
[DistanceFunctionEuclideanDistance]⟹
[10,"AggregationFunction"Mean,"OutlierIdentifier""QuartileIdentifier"]
 | GNNMonUnit |
| GNNMonMakeNearestFunction |
| GNNMonComputeThresholds |
Out[2]=
GNNMon
|
|
Here we find anomalous points in the given set:
In[3]:=
anomalies=gnnObj⟹
⟹
| GNNMonFindAnomalies |
| GNNMonTakeValue |
Out[3]=
{{7.58287,12.4769},{-9.05321,8.66659},{7.3995,13.3589},{-0.603773,18.7801},{-0.596697,2.42631},{-6.93107,6.75308}}
Here we plot the original data together with the found anomalies:
In[4]:=
ListPlot["data"MapIndexed[Callout[#1,#2〚1〛]&,Normal@dsPoints[Values]],"anomalies"anomalies,PlotStyle{{},{Red,PointSize[0.012]}},PlotTheme"Detailed",ImageSizeLarge]
Out[4]=
Contents cannot be rendered at this time; please try again later or download this notebook for full functionality »
Here with find the anomalous points positions:
In[5]:=
gnnObj⟹
[Automatic,"AnomalyPositions"]⟹
| GNNMonFindAnomalies |
| GNNMonTakeValue |
Out[5]=
{13,17,18,19,20,26}
Scope
(1)
 |
|
 |
|
""