Sorry, this algorithm presented by Patricia works, but not always. In fact, if you consider reduced trinomial x^2-7, then her algorithm will find GCF=1, pair of numbers that multiply to get -7 are 1 and -7 or -1 and 7, but what is b?? It is zero here. None of those pairs add up to zero. They add up to 6 or -6. A small trouble happened. Although this trinomial is, in fact, easily factored: x^2-7=(x-sqrt(7))*(x+sqrt(7)).
Basically, we have to use two common situations:
1) If a trinomial is in the form x^2-a^2, it is factored as (x-a)(x+a). Check this yourself by expanding. So if you have x^2-4, it can be written as x^2-2^2=(x-2)(x+2). What if we have x^2-5? It can be written as x^2-(sqrt(5))^2, then it is factored as (x-sqrt(5))*(x+sqrt(5)). Try x^2-169 now. Then x^2-85. If there is any factor in front of x^2, it can be factored out. Example: 2*x^2-200=2*(x^2-100)=2*(x-10)(x+10)--done!
2) A trinomial can be written as x^2+2*b*x+b^2 (Or x^2-2*b*x+b^2) I hope you realize the complete square here. That is, x^2+2*b*x+b^2=(x+b)^2=(x+b)*(x+b). Same is true if there is minus sign in front of 2*b*x term. x^2-2*b*x+b^2=(x-b)(x-b). Example: x^2-6x+9. It can be written as x^2-2*3*x+3^2=(x-3)(x-3). Another example: x^2+10x+25. Figure it out for yourself, I am sure you will do it easily now. If, again, there is any factor in front of x^2, see if the trinomial can be written in the form a^2*x^2+2*a*b*x+b^2 (or a^2*x^2-2*a*b*x+b^2). The rest is the same, it is a complete square and is factored easily.
3) Now in the most common case, when the trinomial is neither in reduced form nor in the form of complete square. Here we have to combine two approaches.
a) First, make a*x^2 a square of some number. It is easy: a*x^2=(sqrt(a)*x)^2.
b) Second, let us create a term 2*sqrt(a)*x. But we have just b*x, you may say. OK, we will do this trick: b*x=2*sqrt(a)/(2*sqrt(a))*b*x. We multiplied and divided b*x by 2*sqrt(a)--we can do it. Let us then recombine the factors here: b*x=[2*sqrt(a)*x]*[b/(2*sqrt(a))].
3) Now, to form a complete square we need a third term, which will be the square of the second factor in square brackets above: [b/(2*sqrt(a)]. So we need b^2/(4a). But we have only c! No problem, we add and subtract that term, b^2/(4a), then we do not change anything and the expression will remain the same. And so we will do. We end up with b^2/(4a)+{c-b^2/(4a)}. Now let us combine everything together.
a*x^2+b*x+c=(sqrt(a)*x)^2+2*[sqrt(a)*x]*[b/(2*sqrt(a))]+b^2/(4a)+{c-b^2/(4a)}. Now we can form a complete square from the first three terms and obtain:
a*x^2+b*x+c=[sqrt(a)*x+b/(2*sqrt(a))]^2+{c-b^2/(4a)}. What is next? OK, let me re-write it a little bit differently: a*x^2+b*x+c=[sqrt(a)*x+b/(2*sqrt(a))]^2-{b^2/(4a)-c}. Do you see anything interesting now?
4) I hope you have already realized something: it is in the form of a reduced trinomial, same as in p. 1)!! We can easily factor it now.
[sqrt(a)*x+b/(2*sqrt(a))]^2-{b^2/(4a)-c}={[sqrt(a)*x+b/(2*sqrt(a))]-sqrt[b^2/(4a)-c]}*{[sqrt(a)*x+b/(2*sqrt(a))]+sqrt[b^2/(4a)-c]}. Looks ugly, right? OK, we will take sqrt(a) factor out from both multiples here, then we will obtain:
a*x^2+b*x+c=a*{x+b/(2a)-sqrt[b^2-4ac]/(2a)}*{x+b/(2a)+sqrt[b^2-4ac]/(2a)}
Our task is done here. Of course, this factoring is possible only if b^2-4ac>=0, which you already recognized as a discriminant from quadratic formula. If b^2-4ac=0, you end up with a complete square (check it yourself!).
Hope it helps!
Patricia S.
With all due respect, Kirill, the example that you gave does not support your statement that my algorithm does not work.
Point #1: You stated: "if you consider reduced trinomial x^2-7, then her algorithm will find GCF=1, pair of numbers that multiply to get 7 are 1 and 7 or -1 and -7, but what is b??"
If you factor out a GCF of 1, your reduced trinomial becomes 1 * (x^2 - 7). Since the 1 is considered trivial by mathematicians because it does not change the meaning of the equation, this would then be written as x^2 - 7, which is exactly where you started.
Secondly, 7 and 1 (and -7 and -1) are TWO pairs of numbers that multiply to 7, but as you pointed out, they are not the only ones, especially if you consider radical numbers.
Point #2: Yes, b = 0 in the example that you cited. You suggested that the factored version of x^2 - 7 = (x-sqrt(7)) * (x+sqrt(7)).
I would like to point out that -sqrt(7) + sqrt(7) = 0. And so your example still supports my algorithm.
08/13/13