Lecture 3: Literature Search

Lecture 3: Literature Search &

Efficient Learning

Elena Dubrova

ELE/EECS/KTH

p. 2 - IL3606 – The Art of Doctoral Research

Overview

• Literature search

• Why to do it

• Where to start

• How not to get lost

• Techniques for efficient reading/learning


Why to do literature search?

• Sometimes a week of reading can save a year of

hard work

• To increase the likelihood that someone will care

about your work – Not duplicate an already known result: “Those who

don’t know the past are condemned to re-discover it.”

• To identify groups of people who will appreciate

what you are doing


Why to do literature search?

• To be able to write a convincing paper:

– Using standard terminology and notations

– Comparing your contribution to related work

– Not getting crushed by some crusty professor

who claims to have published your result

twenty years ago


Cons of literature search

• As a newcomer, you are in a unique position to

see the problem clearly from a fresh perspective

• By reading the literature, you may accept the

prevailing wisdom unquestionably, and

unconsciously close paths of investigation that

might otherwise have occurred to you



• Harish-Chandra (Cole Prize in Algebra, 1954):

“I have often pondered over the roles of

knowledge or experience, on the one hand, and

imagination or intuition, on the other, in the

process of discovery. I believe that there is a

certain fundamental conflict between the two,

and knowledge, by advocating caution, tends to

inhibit the flight of imagination. Therefore, a

certain naivete, unburdened by conventional

wisdom, can sometimes be a positive asset.”



• Albert Einstein (Nobel Prize in Physics, 1921):

”Imagination is more important than knowledge”


Compromise?

• A good strategy when starting research is to give

yourself some time for exploring first, and read

second

– Try to think of as many possible solutions and

approaches as you can

– Then do some literature search. If it’s a problem that

has been encountered before, most probably some of

your ideas have been explored already

• You may need to repeat explore-read cycle many

times – At each cycle you will have a higher level of expertise


Common structure of Publications

• Conference papers

• Workshop papers

• Journal papers

• Technical reports

• Tutorials

• Handbook papers

• Survey papers

• Textbooks and monographs


Specific of publications in our area

• In many science disciplines (physics, biology),

conferences are very informal, papers are

accepted based on abstracts of often incomplete

work

• In my area, conferences: – Many of are highly prestigious with very low

acceptance rates, e.g. ‹ 20% (DAC, ICCAD, DATE)

– Often papers are expected to be highly polished and

complete. Their level is comparable to journal papers.

– Due to space limitations (4-page limit) conference

papers are rarely self-contained


Specific of publications in our area

• In my area, journals:

– Often these are longer versions of papers

previously published at conferences, that

contain more background, details, etc.

– Many researches don’t bother sending to

journals


Where to start

• To search the literature effectively, you have to: – know which questions to ask, and

– use the right terminology in the field

• Identify all the research areas that seem relevant

to your problem

• Look for: – Textbooks

– Tutorials, surveys

– Introductory chapters of theses/dissertations

• These will introduce you to the right terminology


Where to start

• A good strategy is to search for your subject area

and one of these keywords:

– Tutorial, summer school, handbook,

dissertation, thesis, survey, introduction


Things to collect which searching

• As you search the literature, you should be

building sets of:

– Effective search terms and terminology

– Key papers

– Key researchers


Search strategies

• Backward bibliography search – Given a paper in a field, search backward through the

papers it cites

• Forward bibliography search – Given a key paper in a field, search forward to find

papers that cite it

• By researcher – Identify key researchers working in an area. Do

bibliographic search on their names.


How not to get lost

• When you are trying to understand a new area,

nothing will make sense to you. – It is a waste of time to try to read a paper in detail,

you’ll only get frustrated

• Instead, briefly look through several of papers

and try to get a feeling for the material – Look for key concepts and collect keywords

• After a while, the field will gradually come into

focus, and papers that were impossible to

understand before will become more clear


Advices on efficient reading

• Typically, you spend 20% of time to read and

understand 80% of an article/book content and

80% of time to understand the rest

• This 20/80 rule also applies in many other

scenarios


From ” Make the Most of Your Mind”,

T. Buzan, 1984

• Briefly run this your eyes through the pages of an

article/book (as you would do in a bookstore

when you decide whether to buy it or not) to get

an overall impression about 1. its content

2. its structure

3. its level of complexity

4. titles of sections

5. figures and charts

6. conclusions



T. Buzan, 1984

• When you start reading, decide how much time

you will read

• After that, decide which amount of material you

will cover during the allocated time (put a marker

at the beginning and the end of the selected

section) – Determining the end points (in time and space) is

important for maximizing your brain’s efficiently

– ”Cut elephant into small pieces”



T. Buzan, 1984

• Any learning process, including reading, should

be partitioned into intervals of length 20-50 min

with short breaks in between

• Breaks are important because: – They allow you to rest and relax

– The allow the new information which you just read to

be integrated with the information already stored in

your brain



T. Buzan, 1984

• The diagram below shows the amount of material which

is remembered during learning with and without breaks

1h 2h 0h

Amount of remembered

material Learning for 2 hours without breaks

Learning for more than 2 hours without breaks

Learning for 2 hours with short breaks



T. Buzan, 1984

• With time, the amount of remembered material

deceases sharply (after some initial “jump”)

2 days 1 week 1 month 4 months

Amount of remembered

material


To remember for a long time …

• Repeat the material after 2 days, 1 week, 1 month

and 4 months

25%

50%

75%

100%

0% 2 days 1 week 1 month 4 months

1st 3rd 2nd

Repetition intervals 4th

To the long-term

memory



T. Buzan, 1984

• Before starting reading in more details, focus for

a few minutes on the subject and try to

remember what you already know

– This will help you to better ”tune” to the receiving of

the new material and activate your brain into the right

direction

• Next, think for a few minutes on what do you

want to learn from the article/book – Formulate questions which you would like to be

answered



T. Buzan, 1984

• It is OK to skip the hard parts first and return to

them later – We typically assemble a puzzle

from the corners, they are easy

• If you leave a hard part for later, your brain gets

a break during which it will continue working on

the problem on the subconscious level – Have you even been in a situation when

you cannot answer a question which

someone asks you, but later the answer comes

comes to you easily?


NOTE: Learning process is individual

• It is very important to understand how do you

learn in the most efficient way

• How do you learn best?

– Are you visual, audio, physical or verbal learner?

– See a test at http://murdoch-university-futurestudents.blogspot.se/2011/11/are-

you-visual-audio-physical-or-verbal.html

• How do you think best? – Sit quietly and think by yourself

– Talk things over with others see http://www.keirsey.com/4temps/overview_temperaments.asp


Learning process is individual, cont.

• When do you work best? – Do you work better at certain times of the day?

– Our ability to concentrate changes during a day

• There are natural peaks and downs

• More generaly, it is important to understant what

is your strong side – A. Enstein: ”Everybody is a genius, but if you judge a

fish by its ability to climb a tree, it will live its whole life

believing that it is stupid."


Source: doreenchew91tesl.blogspot.se/2013/12/for-fair-selection-everybody-has-to.html

Learning process is individual, cont.

Lecture 3: Literature Search

Documents