Just finished Part 5. It's getting harder but more interesting.
First, in part 5 'amortized analysis' methodology (from Part 4) has been widely adopted serving as the basic tool for Part 5 algorithms complexity analysis.
Chapter 18 is B-Tree which is a generalized tree structure; Chapter 19 is 'Binomial Heap' which is essentially a recursive structure so many algorithms complexity analysis work adopts recursion methodology. In my opinion 'Binomial Heap' is a quite elegant structure because it complies with Recursion perfectly; Chapter 20 'Fibonacci Heap' is the most difficult part in this part - the maths here is much more complex.. Chapter 21 is about set algorithms. The most difficult point here is still the maths for proof of the complexity. Some quite interesting techniques from Computation Theory is introduced here... tough, but really fun.
Just like R-B Tree, I need some time to review the whole stuff for Part 4 & 5.