Getting properties in label 1 that is not exist in label 2

[ibr_ibrahem]

HELLO
I have a csv file with 7 columns i need to get numbers in Made_Call which are not in Receive_call
I had tried many different ways such as [not a.property in b.property] or [none(a.property in b.property where a.property<> b.property)] but always return the common number in the two labels
only apoc subtract function returned the right result but it takes to much time
for larger recordes and also when request to print other related properties gave same result for mentioned ways above .

I need the best way to get the number in column1 that are not in column2
graph relationship based way is prefered

From_Locationa|Made_Call|To_Location|Receiver_Call|Exchange|Duration_of_Call|EVENT_TYPE

06F610044E7D3D	919360123	null	913390956	MSC-BC1 BENGHAZ	00:00:28	Voice
06F610044E7D3D	918856289	null	917078034	MSC-BC1 BENGHAZ	00:01:29	Voice
06F610044E7D3D	911536552	null	7.72E+12	MSC-BC1 BENGHAZ	00:00:28	Voice
06F610044E7D3D	919393414	06F61004507E39	2.19E+11	MSC-BC1 BENGHAZ	00:00:20	Voice
06F610044E7D3D	916176032	null	925742429	MSC-BC1 BENGHAZ	00:00:06	Voice
06F610044E7D3D	916176032	null	925742429	MSC-BC1 BENGHAZ	00:00:06	Voice
06F610044E7D3D	916176032	06F610044F7DF3	125555776	MSC-BC1 BENGHAZ	00:00:06	Voice
06F610044E7D3D	910592535	null	910338043	MSC-BC1 BENGHAZ	00:01:37	Voice
06F610044E7D3D	922151021	06F610076D7AB5	2.19E+11	MSC-BC1 BENGHAZ	00:04:09	Voice
06F610044E7D3D	912003151	null	910338043	MSC-BC1 BENGHAZ	00:00:11	Voice
06F610044E7D3D	912003151	null	6.72E+12	MSC-BC1 BENGHAZ	00:00:11	Voice
06F610044E7D3D	925742429	null	919360123	MSC-BC1 BENGHAZ	00:00:35	SMS
06F610044E7D3D	9176595811	06F61008FD5250	2.19E+11	MSC-BC1 BENGHAZ	00:00:14	Voice
06F610044E7D3D	929100035	null	911439116	MSC-BC1 BENGHAZ	00:00:12	Voice

[arthurnamiasdec]

Hi Ibi
Welcome; let see if we can find an alternative solutions
Rgds
Arthur

[anthapu]

I see one bad usage of relationships here in the model. You have 2 way relationship (Made_Call_To and Received_Call). They are redundant from model perspective unless there is a scenario where only one of them is possible.

Also, if you can post your Cypher query it might be easier to understand what is happening and what needs to be done.

[ibr_ibrahem]

thank you for responding
i know that is not correct but i have tried a lot of relented solutions presented in this and other developers communities without reaching to solution, i uploaded this picture as a sample of the estimated result should be like .

[anthapu]

Can you please post the cypher query to see what is happening?

[ibr_ibrahem]

//9USING PERIODIC COMMIT
LOAD CSV WITH HEADERS FROM "file:/cdr33.csv" AS ROW FIELDTERMINATOR';'
///////////////////////////////////////FIRST PART///////////////////////////////
//////////////////////////////////////CAll NOT RECEIVE//////////////////////////
CREATE (CallinNode:Make_Call {  }) 
SET CallinNode = ROW,
  CallinNode.Duration_of_Call = toint(ROW.Duration_of_Call)
CREATE (CalledNode:Reciever_Call {  }) 
SET CalledNode = ROW,
CalledNode.calledPartyNumber= toInt(ROW.Receiver_Call)
 ;
match(a:Make_Call)
match(b:Reciever_Call)  
where a.EVENT_TYPE = "Voice" 
with apoc.coll.subtract(Collect(a.Made_Call),Collect(a.Receiver_Call)) as  x
Unwind x as k
CREATE (u:Call_Not_Receive { phone: '' })
FOREACH (f IN k | 
    SET u.phone = apoc.convert.toString(f)
)
;
//////////////////////////////////////SECOND PART////////////////////////////////////////
match(a:Make_Call)
match(b:Reciever_Call)
with a as n, b as m
where n.EVENT_TYPE = "SMS"
//where Tower = 1
with apoc.coll.subtract(Collect(n.Receiver_Call), Collect(n.Made_Call))  as c
Unwind c as b

CREATE (p:RECEIVE_SMS_NOT_SENT { phone: '' })
FOREACH (f IN b| 
    SET p.phone = apoc.convert.toString(f)
);

here code that i am using
and what i am looking for another query that gives the same result but graph based