anyone got time? R is a relation from A to B. S and T are relations from B to C. Prove (S o R)\(T o R) ⊆ (S\T) o R
nobody got time for this! Just kiddin!:)
\(S\setminus T \equiv S\cap T^C\)
so are you saying I need to expand (S o R)\(T o R) and then compare with \[(S\cap T^C) o R \] ?
urg... that latex is supposed to show|dw:1397083047858:dw|
First consider what you could do base on the algebraic properties you know. If that doesn't work you look all arbitrary elements.
well, here is what I have so far: S o R = { (a,c) ∈ A x C | ∃b∈B ( (a,b)∈R and (b,c) ∈ S)} T o R = { (a,c) ∈ A x C | ∃b∈B ( (a,b)∈R and (b,c) ∈ T)} (S o R)\(T o R) = {(a,c) | (a,c)∈(S o r) and (a,c)∉(S o T) } I expanded (a,c)∈(S o r) and (a,c)∉(S o T), so ∃b∈B ( (a,b)∈R and (b,c) ∈ S) and ~∃b∈B ( (a,b)∈R and (b,c) ∈ T) then I distribute the negation sign: ∃b∈B ( (a,b)∈R and (b,c) ∈ S) and ∀b∈B ( (a,b)∉R or (b,c) ∉ T)
(S o R)\(T o R) = {(a,c) | (a,c)∈(S o r) and (a,c)∉(S o T) } This doesn't make sense
oh typo O.O I meant, (S o R)\(T o R) = {(a,c) | (a,c)∈(S o R) and (a,c)∉(T o R) }
follow by ∃b∈B ( (a,b)∈R and (b,c) ∈ S) and ~∃b∈B ( (a,b)∈R and (b,c) ∈ T) ∃b∈B ( (a,b)∈R and (b,c) ∈ S) and ∀b∈B ( (a,b)∉R or (b,c) ∉ T)
the first line got cut off
... ~(a,b)∈T and something?
perhaps change from \(\exists b\) to \(\forall b\). That was a mistake
Actually maybe start off from ∃b∈B ( (a,b)∈R and (b,c) ∈ S) and ∀b∈B ( (a,b)∉R or (b,c) ∉ T)
You should define (S\T) o R
ok, that's where I got stuck. I was thinking about using distributive law, but not sure
define (S\T) o R ? ok, hold on
{(a,c} ∈ AxB | ∃b∈B ( (a,b) ∈ R and (b,c)∈(S\T)) }
yes?
you'll have to expand it
ok, hold on...
∃b∈B ( (a,b) ∈ R and (b,c)∈ S\T ) ∃b∈B [ (a,b) ∈ R and ((b,c)∈ S and (b,c)∉T)] Yes?
using associative law, i can remove some parenthesis ∃b∈B ( (a,b) ∈ R and (b,c)∈ S and (b,c)∉T ) yes?
yes
ok, so i have on the left side: ∃b∈B ( (a,b)∈R and (b,c) ∈ S) and ∀b∈B ( (a,b)∉R or (b,c) ∉ T) and on the right side: ∃b∈B ( (a,b) ∈ R and (b,c)∈ S and (b,c)∉T )
ohhh, i think i got something. Wouldn't (a,b)∉R a false condition?
Not necessarily. The \(b\)s are not the same.
but it says all b's are not in R
i'm looking at the statement ∀b∈B ( (a,b)∉R or (b,c) ∉ T)
suppose bo, then (a,bo)∉R or (bo,a)∉T
if we check all element in B, then suppose b1, then (a,b1)∉R or (b2,a)∉T . . . all the way to the last element of B
\[ (a,c) \in (S \circ R)\setminus(T \circ R) \\ \implies \exists b\quad (a,b) \in R\land (b,c)\in S \land \lnot[\exists b'\quad (a,b') \in R\land (b',c)\in T] \\ \implies \exists b\quad (a,b) \in R\land (b,c)\in S \land \forall b'\quad (a,b') \notin R\lor (b',c)\notin T \] We can say :\[ \forall b'\quad (a,b') \notin R\lor (b',c)\notin T\implies (a,b)\notin R\lor (b,c)\notin T \]So we have \[ \exists b\quad (a,b) \in R\land (b,c)\in S \land(a,b)\notin R\lor (b,c)\notin T \]
We distribute across the \(\lor\).
\[ \exists b\quad (a,b) \in R\land (b,c)\in S \land [(a,b)\notin R\lor (b,c)\notin T]\\ \implies \exists b \\ (a,b) \in R\land (b,c)\in S \land(a,b)\notin R \\ \lor\\ (a,b) \in R\land (b,c)\in S\land (b,c)\notin T \]The first one is always false due to \( (a,b) \in R \land(a,b)\notin R\), so it simplifies to: \[ \exists b\quad (a,b) \in R\land (b,c)\in S\land (b,c)\notin T \]
how does \[\forall b'\quad (a,b') \notin R\lor (b',c)\notin T\implies (a,b)\notin R\lor (b,c)\notin T\] ?
Implicitly, \(\forall b'\in B \quad f(b') \implies b\in B \quad f(b)\)
If it is true for every element in B, then it is true for some arbitrary element in B.
Even more explicit, \(\forall b'\in B \quad f(b') \implies \exists b\in B \quad f(b)\)
Join our real-time social learning platform and learn together with your friends!