I would like to get suggestion from you guys that how to sort the fille particularly with given name eg: main engine from remarks and categories the spares used under main engine on remarks and itemname columns
po particular machinery.xlsx (1.3 MB)
thanks you samir my idea s to separate the machinery wise to and to sub catergories what all spare parts been purchased in an year time for particular machinery and need to analysis the price and cost of the particular ship
I have sent an part of the file where it has large numbe rof ship details and different machineries
i need to classify them according to the machineries been used and provid an tag by crating an new column .
i will attach an sample documnt how i have extracted the details
this i have tried using particular filter and created an column as MACHINERY AND CONPONENT
Analyse the price by Vessel is not an issue (you have all you need to do this analysis), so I’ll keep that out.
So, your real issue is to “predict” “Machinery” & “Component” columns from “Remarks” & “ItemName” columns text, yes ? Indeed, you could use ML or DL classification or Text processing as you prefer !
Do you have a list ? A dictionary ?
How do we know that it is an “Auxiliary Engine” or else ?
How do we attribute a component class ? From a word within “ItemName” ?
I think there is a misunderstanding here. I understand you need to create two new columns to identify the “Machinery” and the “Component”.
My question was, “how do we know how to categorize the machinery ?”.
Here is an example for the “Component” column : Predict_Component.knwf (115.1 KB)
I used a Gradient boosted method to identify the Component by reading “Remarks” & “ItemName”.
I reach an 87% success rate, which is not very high. But I think this rate can be improved by using “Text Processing”.
Credits : I used @sjporter “Text processing” component to clean up the text, which is fantastic & easy (so thank you) :