Help with AIM system
|
 |
Overview
AIM is a computer system that assists in the teaching of some parts
of University mathematics. It was developed at the Ghent University in
Belgium. Your comments and suggestions will be very welcome; you can
email
them to us.
The basic idea is that you follow the obvious links from the AIM
home page, and then the system will present you with a list of questions.
There is detailed advice below on how to enter your answers.
Once you give your answers, you can then click the Validate button
at the bottom of the page to mark your work.
Most tests have a "Due date" listed in the test selection menu and at
the top of the test page. In this case you must get them done before the
due date; thereafter your answers will not be recorded. After the due date
the system will tell you your marks and give the solutions to the questions.
The questions are randomly generated, so different students will get
different questions.
How do I enter my answers?
You should type in your answers using the same syntax used in the symbolic
mathematics package Maple. This is broadly similar to the syntax
used for mathematical formulae in general programming languages such as
Java, C and Basic and in spreadsheet programs such as Excel, so you will
find it useful to master it. If you already know Maple, you will find it
easy to enter your answers. If you are not, the advice below should be
sufficient.
When you have entered your answer, you can always click on the Validate
button at the bottom of the page to see how Maple will interpret it. If
there are any syntax errors or Maple has misunderstood what you meant then
you can fix the problem before asking the system to mark your work.
-
Use a star for multiplication: for example, 3x
should be entered as 3*x. Forgetting
this is probably the most common source of syntax errors. Note that you
do not need a semicolon at the end, unlike when you are using Maple
directly.
-
Use a caret (^) for raising something to a power: for example, x2
should be entered as x^2. You can
get a caret by holding down the SHIFT key and pressing the 6 key on most
keyboards.
-
When in doubt, use brackets. For example,
should be entered as (a+b)/(c+d).
If you type a+b/(c+d), then the computer
will think that you mean
If you type (a+b)/c+d, then the computer
will think that you mean
If you type a+b/c+d, then the computer
will think that you mean
Some other examples:
-
2a+b should be entered as 2^(a+b)
-
2 cos 3x should be entered as 2*cos(3*x)
-
eaxsin(bx) should be entered as
exp(a*x)*sin(b*x)
-
(a x2 + b x + c)-1 should
be entered as (a*x^2 + b*x + c)^(-1).
-
Standard functions such as sin, cos, tan, exp, log and so on can be entered
using their usual names. However, the argument must always be enclosed
in brackets: sin x should be entered as sin(x)
and so on. You can use either log(x)
or ln(x) for the natural logarithm
of x. The function 1/sin(x) must be referred
to as csc(x) rather than cosec(x)
(or you can just call it 1/sin(x)
if you prefer). You should always write exp(x)
for ex.
-
sin2x should be entered as sin(x)^2
(which is what it really means, after all). Similarly for tan2(x),
sinh2(x)
and so on.
-
Recall that sin-1(x) traditionally
means the number t such that sin(t) = x, which is of course completely
different from the number sin(x)-1 = 1/sin(x).
This traditional notation is really rather unfortunate and is not used
by Maple; instead, sin-1(x) should
be entered as arcsin(x). Similarly,
tan-1(x)
should be entered as arctan(x) and
so on.
-
Greek letters can be entered using their English names: for example, enter
a+b
as alpha+beta, and 2p
as 2*pi.
-
Vectors can be entered simply as a Maple list. For example, enter the vector
(1,2,3)
as [1,2,3].
-
Matrices can be entered using Maple's "matrix" command. For example, enter
[ 1 2 3 ]
[ ]
[ 4 5 6 ]
as matrix([[1,2,3],[4,5,6]]) or matrix(2,3,[1,2,3,4,5,6]).
Sometimes you will be given the form of a matrix; in this case you can
just fill in its entries.
Sets can be entered with curly brackets: {}. For example enter
[1 0] [1 0] [0
1]
{[ ], [ ],
[ ]}
[0 1] [1 0] [0
1]
as {matrix([[1,0],[0,1]]), matrix([[1,0],[1,0]]),
matrix([[0,1],[0,1]])}.
Penalty for wrong answers
If you press "validate", usually the system will tell you whether your
answers are right or wrong (except possibly for multiple choice questions)
and may give some feedback about any answers that are wrong, but it will
not tell you the right answers. However, penalty (usually 15%) is charged
for each incorrect attempt (except when your answer is syntactically
incorrect or of a wrong type -- in this case a warning will be issued without
any penalty). Thus guessing is counterproductive. Think and try to verify
your answer before entering it. It is better to come back another time
rather than give an answer you are not sure of. For example, if you get
a question wrong three times and then get it right on the fourth attempt,
you will get 55% on that question.
You will not be penalized for syntax errors or type mismatches. Also,
the penalty system keeps track of your answers: even if you have entered
the same wrong answer twice you will be penalized only once.
Minimum number of answers
In some quizzes the teacher may require that a minimum number of questions is answered
before the student obtains some kind of feedback. A question will be considered as answered if it (or one of its subquestions) has been answered syntactically correctly. In that case also the "Validate and show
solutions" changes into a "Validate and show marks" button and only the total mark of the quiz will be
shown (not the results of any individual questions).
Maximum number of attempts
In some quizzes the teacher may impose a maximal number of attempts to submit a webpage.
A counter will then keep track of the number of times that a page has been submitted (via the "Validate" or "Validate and show ..." buttons), no matter whether the page was altered or not.
Hints and subquestions
Sometimes you will get questions with hints. For example:
| Before viewing the hint |
After viewing the hint |
|
|
|
Each hint has a value and a penalty associated to it, as a percentage of
the question value. For the above example the penalty is 20% and the value
is zero. This means that if you gave a correct answer without viewing the
hint, you would get 100% for this question, but if you viewed the hint,
then you would be able to obtain a maximum of 80%.
If you expand a hint whose value is non-zero, then such hint will ask
you a question which you will have to answer. The value of such a subquestion
is expressed as the percentange of the value of the question itself. For
example:
| Before viewing the hint |
After viewing the hint |
|
|
|
For this example:
If you give a correct answer without viewing the hint, you would get 100%
for this question.
If you view the hint and then give a correct answer to both the question
and its hint (as done above), you would get 90%.
If you view the hint, then answer the hint correctly and the question incorrectly,
you would get (0+0.5)/(1+0.5)*90% = 30%.
If you view the hint, then answer the hint incorrectly and the question
correctly, you would get (1+0)/(1+0.5)*90% = 60%.
Subquestions may contain further subhints or questions, in which
case this grading scheme is applied recursively.
Penalties are additive: for example if you use one wrong trial and one
hint with penalty 10% before correctly answering the question, you would
get 100-10-15 = 75%.
Types
Sometimes, the system will check the type of your answer (eg whether it
is a constant, a function of x, a vector, a matrix ...) before working
out whether it is correct. If your answer has the wrong type the system
will allow you to go back and change it without any penalty. Unfortunately,
the warning messages for type errors are not as comprehensible as they
might be. Here are some examples that illustrate how they work.
Suppose that the correct answer is the matrix
[ 1 2 ]
[ ]
[ 3 4 ]
but you enter the number 7 as your answer. You will get the following message:
Your answer has the following error:
Wrong type. Your answer was
7
It should be of type matrix but
it is not.
Try again.
This means that your answer was a matrix, but in fact it is just a
single constant.
Suppose that the correct answer is x2 + y2, but your
answer is x2 + y2 + z2. You might get
the following message:
Your answer has the following error:
Wrong type. Your answer was
x^2 + y^2 + z^2
It should be of type polynom(constant,[x,y])
but it is not.
Try again.
This means that your answer was supposed to be a polynomial function
of x and y, but in fact it depends on z as well. For example, the expressions
x2 + y and (x2 + y2)/6 are polynomials
in x and y, so these answers would have the right type, although they are
not of course the right answer.
Suppose that the correct answer is 2 sin(3 x) but your answer is just the
number 5. You might get te following message:
Your answer has the following error:
Wrong type. Your answer was
5
It should be of type dependent(x) but
it is not. Try again.
This means that your answer was supposed to depend on x.
Multiple response questions
Sometimes you will get a multiple response question, where must
choose all correct answers from a list of n correct and incorrect
answers. For example:
Consider the following matrix:
[cos( q ) -sin( q )]
A = [ ]
[sin( q ) cos( q ) ]
Which of the following properties does this matrix posess?
The grading procedure for multiple response questions usually is:
Grade out of 1 = max(0,(Rs-Ws)/R)
where Rs is the number of right choices and Rs (Ws) is the number of
right (wrong) choices that the student had selected. For the above question,
the right answer is "orthogonal" and "non-singular". If you answer "orthogonal",
"non-singular" and "anti-symmetric" then you would get only 50%. You will
usually get a penalty of 15% for each answer which is not 100% correct.
Since there are 2n possible answers for a multiple response
question, guessing is not to your advantage.
Forbidden words
Suppose that the system asked you to calculate sin(p/4)
(for which the answer is of course 2-1/2). One cunning plan
that might occur to you would be to just enter sin(pi/4)
as your answer, leaving Maple to do all the real work. If this worked,
it would remove the whole point, so the system prevents it. For some questions,
there are certain "forbidden words" that you are not allowed to use as
part of your answer; in the example above, the word "sin" would be forbidden.
If you give an answer involving a forrbidden word, the system will tell
you, and you will be able to go back and change it without any penalty.