Introduction to Computational Thinking – ISC1057

Goals for this lecture:

• Go over the syllabus for the course

• Look at some characteristics of computational thinking
• See the difference between computer hardware and computer software
• Introduce the concept of algorithm
• Watch a video which demonstrates that it’s not so easy to describe even a simple
algorithm precisely
• Compare the work required for a “brute-force” algorithm and a “smarter” algorithm
• Get to know each other
 ISC1057 - Computational Thinking Syllabus

Text: 9 Algorithms That Changed The Future: The Ingenious Ideas That Drive
Today’s Computers
Final Grade Determination. Your grade for the course is determined by class par-
ticipation, homework and group projects. The distribution of grading for the course
is:

• Class Participation - 5%
• Quizzes - 15%
• Homework - 50%
• Midterm Project - 15%
• Final Exam - 15%

The in-class quizzes will be taken using the app Socrative which you will access
through your smart phone. Quizzes will be given almost daily and are typically given
at the end of each lecture. No make-up quizzes are given. For the in-class quiz
grade we average your top 20 quizzes so you are allowed to drop some scores. The
class participation portion of your final grade is determined by the percent of ALL
daily quizzes that you attempt. Your score on the quiz does not affect the class
participation grade; essentially we are using the quiz for attendance.
Assigned homework must be done individually and submitted by the due date unless
you are using your late homework days. You will have a total of 7 late homework
days throughout the semester unless extenuating circumstances arise. Weekly
homework assignments are typically available on Tuesdays and due on Thursday of
the next week. Homework assignments consist of problems similar to the examples
discussed in the lecture and assist you in understanding the material. You will be able
to drop one homework assignment during the semester.
The midterm project is much more involved than homework assignments and serves
to enhance or expand on the material covered in class. Consequently, students will be
expected to work on the project over a longer span of time. For each project, several
choices of topics will be suggested to accommodate the varied interests of students.
Students may work alone or in pairs on the project.
This course is approved to satisfy FSU’s Liberal Studies Quantitative/Logical Thinking
requirement.

Liberal Studies Rule: In order to fulfill this requirement the student must
earn a “C” or better in the course.
 Some quotes about Computational Thinking

“Computational Thinking is the new literacy of the 21st century. The goal is for it
to be a fundamental skill used by everyone in the world by the middle of the 21st
century. Just like reading, writing and arithmetic.”

“Even people who aren’t going into computer science or engineering programs, should
be exposed to computer science. Programming teaches you how to take problems,
break them down into smaller problems, and solve them. Any kind of job that involves
problem solving can benefit from computing. ”

“Computers are extremely talented at performing menial tasks quickly; Computational

Thinkers can leverage that power to find solutions to previously impossible problems.”

“I think everyone should get a little exposure to computer science because it really
forces you to think in a slightly different way, and it’s a skill that you can apply in life
in general.”
What is computational thinking? Who needs it? Why?
Computational Thinking is a problem-solving process that includes the
following characteristics.

Abstraction: Decomposition: Breaking

Generalizing patterns problems into parts

Computational

Thinking

Pattern Recognition: Algorithms: Designing

identifying patterns & implementing methods

and trends to solve problems

Decomposition: Breaking problems into simple parts

“If you can’t solve a problem, then there is an easier

problem you can solve: find it”

George Pólya

subproblem

Complex Problem subproblem

subproblem
1. Make chocolate crust
2. Make mousse and chill
3. Wash & slice fresh strawberries
4. Whip cream
5. Make chocolate hearts
6. Assemble
 Abstraction

• Abstraction is the process of filtering out unnecessary information.

• Abstraction allows us to create a general idea of what the problem

is and how to solve it.
Example: Abstracting a general procedure for baking cookies

1. Bring butter and eggs to room temperature

2. Sift dry ingredients
3. Cream sugars and butter
4. Add eggs and vanilla
5. Mix in dry ingredients
6. Mix in add-ins like chips, nuts, etc.
7. Bake
 Pattern Recognition

• The goal of pattern recognition is to find common similarities and

differences among objects.

• This allows us to solve seemingly diverse problems by a single algo-

rithm.
Example of pattern recognition - Putting together Ikea furniture
 Designing algorithms

• Determine the right steps

• Determine the correct order of the steps

• Follow the steps completely

• Verify that the algorithm is working correctly

• Determine if algorithm is the “best” approach for solving problem

Example. Computer algorithm for multiplying a number by an integer
using only addition

Assume that the computer only has the ability to do addition and we
want to multiply something like
351.6 × 14

Let a be a given number which we want to multiply by an integer b

product = 0
for i = 1, b
product = product + a
end
This is called pseudo code for our algorithm because it does not use
language specific terms.
 Computer Hardware vs Computer Software

Computer hardware is any physical device used in or with your computer.

This includes your monitor, mouse, keyboard, hard drive, sound card,
video card, etc.
Computer software is a general term used to describe a collection of
computer programs that perform some task on a computer system.
Computer software is generally divided into two classes – system soft-
ware and application software. Operating systems, utilities, device
drivers, programming languages, interpreters are examples of system
software. Application software are a group of computer software that
are created to help the user complete tasks such as creating documents,
performing calculations, storing and managing data, playing videos, etc.
This includes programs like Microsoft Word, Adobe Acrobat, etc.
Computer hardware operates under the control of software.
We are interested in the algorithms that are contained within the soft-
ware.
What is an algorithm?

“An algorithm is like a recipe, it must be followed exactly, it must be

unambigious, and it must have an end.”

An algorithm is a precise sequence of steps to solve a problem.

An algorithm should generate a solution. The solution may NOT be

the “best” solution to the problem. The algorithm may NOT be the
most efficient way to solve the problem.

An issue that arises in designing an algorithm is that we have to describe

the algorithm in an orderly and unambiguous way. This is not as easy
as it may seem.

As an example, consider the following video from an introductory com-

puter science course at Harvard where the class is attempting to give
three volunteers an algorithm for making a peanut butter and jelly
sandwich. Note that the center participant has a different type of jelly
container than the other two. This is to illustrate that we often have
to take care of special cases when designing an algorithm.
 No Man’s Sky: A Game Crafted by Algorithms

The virtual universe in this program is generated by an algorithm, not

by an artist’s pen. The generated universe is, of course, not infinite but
“if you were to visit one virtual planet every second then our own sun
will have died before you have seen them all.”
The most common example of this type of game is Minecraft which
creates a unique world for each of its players, randomly arranging rocks
and lakes from a limited palette of bricks whenever someone begins a
new game. But No Man’s Sky is far more complex and sophisticated
because the algorithm contains rules for the physics, biology, etc. of
each virtually generated planet. The tens of millions of planets that
comprise the universe are all unique. Each is generated when a player
discovers it, and is subject to the laws of its respective solar systems
and vulnerable to natural erosion. The multitude of creatures that
inhabit the universe dynamically breed and genetically mutate as time
progresses.
 A simple, but outdated, example to compare Brute-Force
and Divide and Conquer algorithms

Consider the problem of looking up

a name in the phone book. (I know
that you have probably never had
to do this in your life!)

If you are looking up, e.g., Mike Shephard in the phone book then the
following would be an algorithm for finding the phone number.
Step 1 Open the phone book to the first page of names, i.e., the ones
beginning with “A”;
Step 2 Check to see if the name Mike Shephard is on the page;
Step 3 If the desired name is not on the page then go to the next page
and go back to Step 2;
If the desired name is on the page then write down the phone
number and stop because you are done.

If you implement this algorithm will you always find the desired phone
number? (Assuming Mike Shephard is listed in the book.)

The way we have written this algorithm is called pseudo-code because

it describes the necessary steps in an ordered and unambigious way
without the use of any programming language.

Notice that we have a conditional, i.e., an if-then statement which

allows us to determine if we are done or if we have to go to the next
page of the phone book.

Note that we also have a repeated loop, i.e., we keep doing steps 2 &
3 until we are done.
Now let’s look at how much work it takes to find the name Mike Shep-
hard using this algorithm.

• Assume that names beginning with “Sh” appear 81% into the book.
This means that if the phone book has 100 pages then “Shepard”
will appear on page 81.
• If the phone book has 200 pages then ‘Shepard” will appear on page
162 = .81 × 200 so we have to look at twice as many pages as when
the phone book had 100 pages.
• If the phone book has 500 pages then ‘Shepard” will appear on page
405 = .81 × 500 so we have to look at five times as many pages as
when the phone book had 100 pages.
• For obvious reasons, this is called a Brute-Force algorithm or the
Exhaustive search algorithm.
• If the phone book doubles in size the next year to 1000 pages and
 “Sh” still appears 81% into the book, we have to look at 810 pages
before we find the phone number. If the book doubles again to 2000
pages then we have to look at 1,620 pages.
• This means that our work (i.e., the number of pages we must scru-
tinize) doubles as the size of the book doubles. This means that if
we represent the work in a graph, we will have a straight line. For
example, if x is the number of pages in the phone book and y rep-
resents the number of pages we have to look at then we write the
line which passes through any of the two points, say (500, 405) and
(1000, 810) so we have
810 − 405 405
y − 810 = (x − 1000) =⇒ y = 810 + (x − 1000)
1000 − 500 500
or
y = 810 + .81(x − 1000)
If there are only 100 pages in the book then y = 810+.81(100−1000) =
810 − .81(900) = 810 − 729 = 81 which agrees with what we said the
amount of work was for a 100 page phone book .

If we are looking for the phone number for Karen Allen then it will take
less work using this algorithm but if we are looking for Eric Yeager then
it will take more work.

In general, the amount of work to find a phone number is proportional

to the number of pages N in the phone book which means it is a
constant times N .
 Work for Brute-Force Algorithm

amount of work

size of problem
What can we do to reduce the amount of work?

Consider the following algorithm which intuitively seems to cut the work
in half.

Step 1 Open the phone book to the first page of names, i.e., the ones
beginning with “A”;
Step 2 Check to see if the name Mike Shephard is on the page;
Step 3 If the desired name is not on the page then skip next page, go
to next page and then go back to Step 2;
If the desired name is on the page, write down the phone number
and stop because you are done.

Does it always work?

 Divide and Conquer Approach

Step 1 Open the phone book to the middle;

Step 2 Check to see if the name Mike Shephard is on the page;
Step 3 If the desired name is not on the page then
• if the names on the page begin with a letter combination before
“Sh” then throw the first half of the book away and go back to
Step 1;
 • if the names on the page begin with a letter combination after
“Sh” then throw the second half of the book away and go back to
Step 1;
• if the desired name is on the page, write down the phone number
and stop because you are done.

How much work does this algorithm take?

Assume again that Shepard appears 81% of the way into a 500 page
book, i.e., Shepard appears on page 405.
 pages pages
Step 1 discard
1-250 251-500
keep

pages pages
Step 2 discard
251-375 376-500
keep

pages pages
Step 3 keep
376-438 439-500
discard

pages pages
Step 4 keep
376-407 408-438
discard
Continuing in this manner, we narrow the search to the following num-
ber of pages
Step 5 376-391 392-407

Step 6 392-399 400-407

Step 7 400-403 404-407

Step 8 404-405 406-407

Step 9 404 405

Now what happens if the phone book doubles in size to 1000 pages and
the phone number is on page 810? Remember before that the work
doubled with the brute-force approach.
Step 1 1-500 501-1000

Step 2 501-750 751-1000

Step 3 751-876 877-1000

Step 4 751-813 814-876

Step 5 751- 782 783-813

Step 6 783- 798 799-813

Step 7 799-806 807-813

Step 8 807-810 811-813

Step 9 807- 808 809-810

 Step # # number of pages to search
500 pages 1000 pages 2000 pages
1 250 500 1000
2 125 250 500
3 62 125 250
4 31 62 125
5 15 31 62
6 7 15 31
7 3 7 15
8 1 3 7
9 1 3
10 1

It turns out that the work required depends upon log N where N is the
size of the phone book. Now it’s not important whether you remem-
ber what a logarithm is, but it is important that you understand the
following plot where we plot the amount of work as the phone book
size increases for both the Brute-Force approach and the Divide and
Conquer approach. For any N > 1 log N is always smaller than N as
can be seen from the plot.
amount of work

e
rc
Fo
e
ut
Br

uer
Divide & Conq

size of problem
 Computer Algorithms are Changing the World

• Information at our fingertips through online searches.

• Buying products securely online.
• Finding “significant others” through online matching services.
• Immediate news coverage around the world
• On demand entertainment.
• Virtual currency

How do you think algorithms & computers will change the the way we
live our lives in the next decades?
Reading Assignment:

1. Computational Thinking, Jeannette Wing1, 2006

2. The Algorithm: Idiom of Modern Science, Bernard Chazelle, intro-
duction and section “Thinking algorithmically”

Both articles are posted on our Blackboard site.

Next time we will have a Socrative quiz so be sure and bring your
phones!

Let’s adjourn for some cookies and get to know each other!

1 Former Chairman of Computer Science at Carnegie Mellon University and currently Corporate Vice President of Microsoft Research
 Introduction to Computational Thinking - Continued

Goals for this lecture:

1. To look at different ways to design algorithms through specific ex-

amples
2. To see how to use the free online plotting program PLOTLY.
3. To understand that two key components of many algorithms are (i)
repeated loops (recursion) and (ii) conditionals (if - then statements)
4. Introduction to Socrative through quizzes
5. Overview of topics we plan to cover this semester
Last time we said that different algorithms for the same problem may
give different answers. The following example illustrates this.

Example. Suppose you are packing your backpack for a hike but you
have 6 items you are interested in taking but their total weight is 96 lbs.
The maximum weight you are willing to carry is 50 lbs. You assign each
item an importance value between 0 and 25; the higher the number the
more valuable. What items should you take?
 Item A B C D E F
weight 30 28 15 10 9 5
value 16 25 24 15 12 2

For the first algorithm we simply pick an item at random and put in the
backpack. We then weigh the backpack and if it weighs < 50 lbs then
we pick another item. If it weighs more than 50 lbs then we remove
the last item and try another item at random.
Algorithm I - Random

Step 0 Put all items in a selection pile.

Step 1 If there are any items in the selection pile, pick an item at
random from the selection pile and put into the backpack. If there
are no items left in selection pile you are done.
Step 2 Weigh the backpack.
If it weighs < 50 lbs, go to Step 1.
If it weights 50 lbs, you are done.
If it weighs > 50 lbs, remove item from backpack and from selection
pile. Return to Step 1.

The strategy behind this algorithm is randomness. It can generate any

possible solution if we perform it many times.
 Items Weight Value
A, C, F 30 +15+5=50 lbs 16+24+2=42
A, D, E 30+10+9=49 lbs 16+15+12=43
A, D, F 30+10+5=45 lbs 16+15+ 2=33
B, D, E 28+10+9 =47lbs 25+15+12=52
B, D, F 28+10+5 =43lbs 25+15+ 2=42
B, C, F 28+15+5=28lbs 25+24+2=51
C, D, E ,F 15+10+9+5 =39lbs 24+15+12+2=53
The strategy behind the next algorithm is to pick the items in their
order from most valuable to least valuable.

Algorithm II - Arrange items from most valuable to least valuable

and add items in order if they fit

Step 0 Arrange the items in a selection pile in order of value from

largest to smallest
Step 1 Pick the next item in order from the selection pile and put into
the backpack. If there are no items left in selection pile you are done.
Step 2 Weigh the backpack.
If it weighs < 50 lbs, go to Step 1.
If it weights 50 lbs, you are done.
If it weighs > 50 lbs, remove item from backpack and the selection
pile; return to Step 1.
What result does this strategy give us?

Step 0 We order our objects as B, C, A, D, E, F

Step 1 Put item B in backpack
Step 2 Backpack weighs 28 lbs < 50 lbs so go to Step 1
Step 1 Pick item C from pile and put in backpack
Step 2 Backpack weights 43 lbs < 50 lbs so go to Step 1
Step 1 Pick item A from pile and put in backpack
Step 2 Backpack weights 73 lbs > 50 so remove item A from backpack
and from selection pile. Go to Step 1
Step 1 Pick item D from pile and put in backpack
Step 2 Backpack weights 53 lbs > 50 so remove item D from backpack
and from selection pile. Go to Step 1
Step 1 Pick item E from pile and put in backpack
Step 2 Backpack weights 55 lbs > 50 so remove item E from backpack
and from selection pile. Go to Step 1
Step 1 Pick item F from pile and put in backpack
Step 2 Backpack weights 48 lbs < 50 so go to Step 1
Step 1 No items left in selection pile so we are done.

Answer: Items B, C, and F which weigh a total of 48 lbs (28+15+5)

and have a value of 25+24+2=51
The strategy behind this algorithm is to order the items based upon
their value per weight so that if you have two items that weigh 10 lbs
and one has a value of 15 and the other of 5, then clearly the item of
value 15 is a better choice for the given weight.
Algorithm III- Compute the ratio of value per unit size and add items
in order if they fit

Step 0 Compute the ratio of value/weight for each item. Arrange the
items in a selection pile in order of value from largest to smallest
ratios.
Step 1 Pick the next item in order from the selection pile and put into
the backpack. If there are no items left in selection pile you are done.
Step 2 Weigh the backpack.
If it weighs < 50 lbs, go to Step 1.
If it weights 50 lbs, you are done.
 If it weighs > 50 lbs, remove item from backpack and selection pile
and return to Step 1.

What result does this strategy give us?

Step 0 We compute the value per unit size of each item as

16 8 25 24 9
A: = B: C: =1
30 15 28 15 15
15 1 12 1 2
D: =1 E: =1 F:
10 2 9 3 5
We order our objects as C, D, E, B, A, F
Step 1 Put item C in backpack
Step 2 Backpack weighs 15 lbs so go to Step 1
Step 1 Pick item D from pile and put in backpack
Step 2 Backpack weights 25 lbs so go to Step 1
Step 1 Pick item E from pile and put in backpack
Step 2 Backpack weights 34 lbs so go to Step 1
Step 1 Pick item B from pile and put in backpack
Step 2 Backpack weights 52 lbs so remove item B from pile and go to
Step 1
Step 1 Pick item A from pile and put in backpack
Step 2 Backpack weights 64 lbs so remove item A from pile and go to
Step 1
Step 1 Pick item F from pile and put in backpack
Step 2 Backpack weights 39 lbs

Answer: Items C, D, E and F which weigh a total of 39 lbs and have a

value of 24+15+12+2 = 53
Which algorithm gives the best answer?
Which algorithm may give the best answer or it may give the worst
answer?
Last time we saw an example of a brute-force algorithm and a Divide
& Conquer algorithm for finding a phone number in a telephone book.

As another example, consider an algorithm for counting; specifically we

will determine the number of people in a room.
The brute-force approach is for me to point at each person and count
1,2,3,4,. . .. This approach requires me to go through all the numbers
from 1 to N , where N is the number of people in the room.

A second, but slightly more efficient, brute-force approach is to count

individuals by two, i.e., 2, 4, 6, 8, . . .. This approach takes half of the
work of counting each person individually.

A smarter approach is the following algorithm.

Step 1 Each person in the room stands up and is assigned the number
1.
Step 2 Each standing person must pair up with another person standing
in the room and add your respective numbers together.
Step 3 One of the two of you must sit down.
Step 4 If more than one person is standing, go to Step 2 and repeat.
If only one person is standing then their number is the total number
of people in the room.

Again, our algorithm has a conditional (an IF-THEN statement) and a

repeated loop.

Let’s try the algorithm to count the number of people (excluding in-
structor) in the class.
This algorithm does not require the room to have an even number of
people in it as the following schematic illustrates with 21 people. Here
it takes 5 loops (or recursive steps) to compute the algorithm.
Step 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Step 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1

Step 3 2 2 2 2 2 2 2 2 2 2 1

Step 2 2 2 2 2 2 2 2 2 2 2 1

4 4 4 4 4 4 4 4 4 4 1

Step 3 4 4 4 4 4 1

Step 2 4 4 4 4 4 1

8 8 8 8 5 5

Step 3 8 8 5

Step 2 8 8 5

16 16 5

Step 3 16 5

Step 2 16 5

21 21

Step 3 21

• At the end of the first recursive step (Step 3) we have 11 people

 standing.
• At the end of the second recursive step (Step 3) we have 6 people
standing.
• At the end of the third recursive step (Step 3) we have 3 people
standing.
• At the end of the fourth recursive step (Step 3) we have 2 people
standing.
• At the end of the fifth recursive step (Step 3) we have 1 person
standing so we are done.
If we double the number of people in the room how many more steps
does it take?
Step 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Step 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Step 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

At the end of the first recursive loop there are 21 people standing which
is the same number as in the previous example which took 5 recursive
steps so when we start with 42 people it takes one more step or 6 total.

The table below compares the number of people remaining after Step
3 is completed (when one member of each pair sits down) for different
size classrooms. Note that as in the Divide & Conquer algorithm the
work increases by one as the number of people is doubled.

Loop # # people standing after each loop

class: 21 class: 42 class: 84 class: 168
1 11 21 42 84
2 6 11 21 42
3 3 6 11 21
4 2 3 6 11
5 1 2 3 6
6 1 2 3
7 1 2
8 1

Based upon the trend we see from this table we can predict how many
steps it will take for a classroom of size 336, 672, etc. and when we
plot the points we get a curve that looks like the log plot we had before.

amount of work

e
rc
Fo
e
ut
Br

uer
Divide & Conq

size of problem
Class Size Number of loops required
21 5
42 6
84 7
168 8
336 9
672 10
1344 11
2688 12
Many people believe that computers are so powerful that they can solve
any problem by a Brute-Force approach. This is not true! We need to
think in different ways to design more efficient algorithms so results can
be obtained in a reasonable amount of time.
In your homework, you are asked to make a similar plot using the
computer (not by hand).
How can you do this? You can use software such as Microsoft or a free
online program called PLOTLY.
I will show you how to create such a plot now using PLOTLY. Simply
do a search for PLOTLY. You must create your own account but it’s
FREE. To turn in your plot you can either save it in the program or
perform a ScreenShot of the plot.
In these algorithms, as well as the ones from last time, we had a con-
ditional statement, i.e., an if-then statement. Here is Bill Gates (Mi-
crosoft) explaining these.

YouTube video www.youtube.com/watch?v=m2Ux2Pnje6E

In these algorithms, as well as the ones from last time, we had a repeat
loop. Here is Mark Zuckerberg (Facebook) explaining the importance
of these.

YouTube video www.youtube.com/watch?v=mgooqyWMTxk

 Socrative Practice Quiz
CTISC1057

Go to b.socrative.com and you will see a login page.

Under Student Login enter the Room Name CTISC1057.
The next page has you enter a name to identify yourself. Use YOUR
FSU LOGIN (such as mc12s; do not include “@ myfsu.edu”)
The first question on the correct quiz should appear. Because the
question is True/False you simply click the correct box.

1. The term “algorithm” is only used to refer to computer programs.

2. Your computer’s trackpad is an example of software.
3. An example of a brute-force algorithm for counting the number of
people in a room is counting the individuals by twos.
4. If a phone book has 1000 pages then the Divide & Conquer algorithm
for finding a number in this phone book takes twice as much work
as for a phone book with 500 pages.
5. Jeannette Wing believes that computational thinking should only be
taught in college.
6. All algorithms for a specific problem give the same answer.
If we want to implement an algorithm on a computer, how is this ac-
complished?

Suppose you have a recipe for hush

puppies and you want to tell a per-
son in China how to make this dish
but they don’t speak English. The
only want to achieve this is to write
the recipe in English and have an interpreter translate it into Chinese.
In this analogy the recipe is the algorithm, you are the programmer and
the person in China plays the role of the computer. You might be able
to write the recipe but the “computer” can’t understand it!
This analogy is like the programmer/computer situation. The program-
mer uses a programming language such as C++, Java, Python, For-
tran, etc.. to write the software. Unfortunately, the computer doesn’t
understand these languages because all it understands is machine lan-
guage. To complicate matters, programmers don’t understand machine
language.
To solve this problem we need the equivalent of an interpreter which
is what the compiler does. It converts our computer program written
in a programming language that we can understand into one that a
computer understands, basically into binary which is a language that
only consists of ones and zeros (we will talk about this later).
Why can’t we write our computer programs in English? Our language
is too imprecise for computers!
We will NOT learn a programming language in this course but rather
concentrate on the ideas behind the algorithms which a programmer
implements.
 Software Bugs

Oftentimes one hears the term bug when referring to software.

A software bug is an error or flaw in a computer program that causes
the result to be incorrect or to behave in unintended ways. These errors
typically occur due to mistakes made by humans either in the actual
program or in the algorithm design.
A study commissioned by the US Department of Commerce’s National
Institute of Standards and Technology in 2002 concluded that ”software
bugs, or errors, are so prevalent and so detrimental that they cost the
US economy an estimated $59 billion annually, or about 0.6 percent of
the gross domestic product”.
Where did the term “bug” come from?
Grace Hopper, a computer scientist and Navy Rear Admiral, joined the
Harvard Faculty at the Computation Laboratory after WWII to continue
her work on the Mark II computer. Operators traced an error in the
Mark II to a moth trapped in a relay, coining the term bug. This bug
was carefully removed and taped to the log book. Stemming from the
first bug, today we call errors or glitches in a program a bug.
Typos in a program are usually caught by a compiler . Computer bugs
often occur when the programmer makes a logic mistake or fails to take
into account special cases (as in the PB&J video with the differences
in the jam containers).
 Computers Are Reshaping Our World

This class looks at ways in which computers have begun to change the
way we live.
Many recent discoveries and inventions can be attributed to the power
of computer hardware (the machines), the development of computer
programs (the instructions we give the machines), and more impor-
tantly to the ability to think about problems in new ways that make
it possible to solve them with computers, something we call computa-
tional thinking.
 Issues in Computational Thinking

What problems can computers help us with? What is different about

how a computer seeks a solution? Why are computer solutions often
not quite the same as what a human would have found?
How do computer programmers write out instructions that tell a com-
puter how to solve a problem, what problem to solve, and how to report
the answer?
How does a computer “think”? How does it receive instructions, store
information in a memory, represent and manipulate the numbers and
words and pictures and ideas that we refer to as we think about a
problem?
We will see that making a computer carry out a task in a few sec-
onds actually can take years of thinking, research, and experiment. So
what seem like simple tasks actually can hide a great deal of technical
construction.
But the rewards can be enormous; depending on what you want, you
can hope that a computational approach will give you an unbeatable
chess player, a drug that protects against the Zika virus, or a car that
safely drives itself to any destination you want.
 Topics to be covered this semester

Google Books & Google Viewer

• One week
• In 2004, Google announced a plan to systematically scan a copy of
every book ever published, estimated to number 130 million. The
result will be a huge database called Google Books.
• The idea was not to just copy all the publications but to make a
searchable data base. To allow everyone easy access to this data
base, Google developed a viewer which allows the user to compare
the number of occurrences of words or phrases in print through the
years.
• We will be looking at some of the history of digitizing text, the
problems Google encountered with this project, the type of computer
algorithms needed to convert text into a searchable document and
how to glean information from the viewer.
Cryptography, Public Key Encryption & Digital Signatures

• 3.5 weeks
• Algorithm # 4 (Public Key Encryption) and Algorithm # 9 (Digital
Signatures) in text
• We will begin by looking at cryptography from a historical perspec-
tive. We will see how it has been used to send secret messages as
early as Julius Caesar (who has a cipher named for him) up through
World War II. You will learn how to encrypt/decrypt a message given
a key and how to decrypt certain types of ciphers without a key by
looking at letter and word frequency plots.
• Cryptography became critical for everyone, not just military leaders,
when computers became prevalent. Now we enter our credit card
information in a website to make purchases with confidence that the
information is secure. How does this work? Public Key Encryption is
 the algorithm that makes this possible. We will strive to understand
this algorithm by looking at simple examples.
• Another application of encryption is when you sign a document elec-
tronically. How can the individual receiving your signed document
know that it was really you who signed it?

GHIHQG WJH HDVW ZDOO RI WKH FDVWOH

G H I H Q G W J H H D VW

D E F E N D T H E E A S T

DEFEND THE EAST WALL OF THE CASTLE

Machine Learning: Linear Regression and Pattern Recognition

• 3 weeks
• Algorithm # 6 in text
• Since the advent of computers, people have considered whether com-
puters can become the dominant form of intelligence on Earth; this
has been a theme in many science fiction stories/movies through the
decades. There is actually a test called the Turing test (1950) which
is used to measure a machine’s ability to exhibit intelligent behavior.
• Machine learning algorithms are used in many applications from read-
ing postal codes to autonomous land vehicles. Unlike other algo-
rithms we have considered, in machine learning we typically “train”
the algorithm with a set of data and then, after it has “learned”,
we input new data for it to identify. For example, when reading
hand written postal codes we could train the algorithm with a set of
 handwritten numbers and then give it a new code to identify.
• Recognizing patterns is something that humans do better than com-
puters. Typically a human can identify an image of a person after
having seen another picture of the person but this ability is difficult
to incorporate into an algorithm.
• We will look at various types of Machine learning algorithms and
see the types of applications they are best suited for. An underlying
theme is there is not one algorithm which works for all problems.
 “1”

“2”
Images of
Machine Learning
Handwritten “3”
Algorithm
Numbers
“4”

etc.

Training an algorithm to read a zip code

Machine Learning - Clustering

• 1.5 weeks
• Clustering is the process of organizing objects or people into groups.
You could say that your friends on Facebook is a cluster and my
friends are another cluster. Your neighbors form one cluster and
my neighbors form another. Clustering is used in many applications
without us even realizing it. For example, if we compare a photo on
our computer with one that we print, we see that they are not exactly
the same because the monitor has many more colors available than
most printers. How did the printer assign a color to a pixel?
• Of course to develop an algorithm to cluster objects requires having
criteria for forming the clusters such as distance, color, etc. We will
look at various types of clustering and applications.
• We will see how clustering was used to discover that the disease
cholera was transmitted through the water.
How can computers solve word games?

• 1 week
• Lewis Carroll (Alice in Wonderland) developed a word game called
Doublets (also called Word Ladder). For example, can you take the
word MAN and turn it into APE by changing only one letter at a
time. Moreover, we would like to do this in the shortest number of
moves possible.
• We will look at strategies for solving this puzzle and variants and
see how they can be implemented in an algorithm for a computer to
solve the puzzle.
• This will give you an insight into algorithms for game solving such as
chess. In 1997 IBM’s chess computer beat the top-rated chess player
in the world and over the next few years humans and computers
traded blows until eventually by 2005-6, computer chess programs
were in the lead.
Sorting items and Searching the Web

• 2.5 weeks
• Algorithm # 2 (Page Match) and Algorithm # 3 (Page Rank) in
text
• Whenever you have items which you want to search repeatedly (such
as web pages), then it is important to first sort the items. We will
look at several approaches to sorting.
• When you do a search on the Web then not only does the search
engine find information that matches your search words but it also
ranks them, i.e., it puts the most relevant matches on the first page.
Moreover, it does this in seconds not hours.
• Just finding matches seems like an impossible task because it is es-
timated that there are over a billion websites. What we will see is
that when you do a search, the search engine doesn’t really search
 the Web but rather a sorted index of the Web which it has created
earlier.
• Ranking the sites which match the search words seems like another
impossible task because a computer doesn’t really understand what
is written on the website. How can it distinguish between spam
websites and ones which have meaningful information? How can it
decide which are the most relevant sites? What can webmasters do
to help their sites appear on the first page of results? These are some
of the questions that we will look at.
Data Compression

• 1 week
• Algorithm # 7 in text
• Big data is the new buzz word for this century. Jobs requiring big
data expertise are predicted to be numerous and carry high salaries.
• Amazon gets 35 orders per second or 1.1 billion per year. That is a
lot of data! In order to handle this much data it must be compressed
in such a way that it can be decompressed at a later date. We want
to look at some of the ideas behind data compression.
Reading Assignment

Computational thinking, 10 years later, Jeannette Wing, Microsoft Re-

search Blog, 2016
 Socrative Quiz Introduction Quiz1
CTISC1057

amount of work

size of problem

1. In the graph of the work required to find a name in a phone book,

 the algorithm described by the blue curve requires less work for a
given phone book than the one described by the red curve.
2. In the graph of the work required to find a name in a phone book
using the Divide and Conquer and Brute-Force algorithms, the red
line is for the Divide and Conquer algorithm and the blue curve is for
the Brute-Force algorithm.
3. Different algorithms for a specific problem may give different answers.
4. At each step of a Divide & Conquer algorithm you break the problem
into smaller pieces.
5. In Jeannette Wing’s article on Computational Thinking she advocates
teaching computational thinking in K-12 grades.
6. In programming lingo a conditional statement means that you have
an If - then statement.
7. The two most common components of an algorithm are repeated
 loops and conditionals.
8. Computer algorithms are better at repeated tasks than humans.
9. Microsoft Word is an example of computer hardware.
10. Computational thinking involves abstraction and breaking complex
problems into smaller parts.