Unique ID's from Data Set

Hi tbhatia ,

You can try to deduplicate the names by grouping by ID, and keep the ones greater than 2.

The complete M language is as follows

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjQwMFTSUXLyD1GK1YFzPUpzE/OKYSJGGCLGMC2xAA==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"ID's" = _t, names = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"ID's", Int64.Type}, {"names", type text}}),
    #"Grouped Rows" = Table.Group(#"Changed Type", {"ID's"}, {{"Count", each _, type table [#"ID's"=nullable number, names=nullable text]}}),
    #"Added Custom" = Table.AddColumn(#"Grouped Rows", "Custom", each Table.RowCount(Table.Distinct([Count]))),
    #"Expanded Count" = Table.ExpandTableColumn(#"Added Custom", "Count", {"names"}, {"names"})
in
    #"Expanded Count"

Initial data

1.Group by and select all rows.

2.Add a custom column.

3.Expand the column and filter out rows with custom column value equal to 2.

Best Regards,

Stephen Tao

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.