1. 

note that the join through section is not needed in the queries
below.  Instead of x.docid = y.docid and y.docid = z.docid, it
suffices to say x.docid = z.docid.

a. 

select x.docid, x.title, count(distinct z.word)
from doc x, section y, wordocc z
where x.docid=y.docid and y.docid=z.docid and y.secnumber=z.secnumber
group by x.docid, x.title

b.

selec distinct x.docid, x.title
form doc x, section y, wordocc z1, wordocc z2
where x.docid = y.docid
  and y.docid = z1.docid and y.secnumber = z1.secnumber
  and y.docid = z2.docid and y.secnumber = z2.secnumber
  and z1.word = 'midterm' and z2.word = 'solution'

b.

select distinct y.docid
from doc x
where not exists ( select * 
                   from QW w
                   where not exists (select * 
                                     from section y, wordocc z
                                     where x.docid = y.docid
                                       and z.docid = y.docid
                                       and z.secnumber = y.secnumber
                                       and z.word = w.word))


2. b

Decompose R(A,B,C,D,E,F,G).  Try A+ = AB.  Decompose into R1(A,B),
R2(A,C,D,E,F,G)

Decompose R2(A,C,D,E,F,G).  Try C+ = ACD.  Decompose into R3(A,C,D),
R4(C, E, F, G)

Decompose R4(C,E,F,G).  Try EF+ = CEF.  Decompose into R5(C,E,F) and
R6(E,F,G)

End result: R1(A,B) (key=A), R3(A,C,D) (key=C), R5(C,E,F) (key=E), and
R6(E,F,G) (key=EFG)

c. Write X<-->Y to mean X-->Y and Y-->X

C-->A, D-->B, AB<-->E
BC--> all attributes
ED--> all attributes
CD--> all attributes

3.a.

(i) T2, T5
(ii) T3, T4, T6
(iii)

     Redo         Undo
A=          | 2
B= 20	    |
C= 	    | 100
D= 2000	    | 2000
E=          |

b. yes, no, no, yes