Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Metric Learning
It is closely related toregressionand classifcaton, but the goal is to learn from examples a similarity functon that
measures how similar or related two objects are.
Large Margin Nearest Neibhour LMNN
It classifes a new data instance with
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Sarcasm Detection
An Obvious Solution to a trivial
problem
Ishan Chutani
Why a Sarcasm detector
A step closer to development of machines
towards a humanoid
Sentiment analysis
Automating everything literally
Possible approaches
Traditional approach
B
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
How do we show/ demonstrate?
Use the predictive maintenance template and convert the bearing data into the template
data format through Matlab. Then we should be easily able to apply the template by writing
some python/R code on the ML Studio.
In Azure, T
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Efficient Learning of Mahalanobis Metrics for Ranking
Daryl K. H. Lim
DKLIM @ UCSD . EDU
Department of Electrical and Computer Engineering, University of California, San Diego, CA 92093 USA
Gert Lanckriet
GERT @ ECE . UCSD . EDU
Department of Electrical a
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Large Margin MultiTask Metric Learning
Kilian Q. Weinberger
Department of Computer Science and Engineering
Washington University in St. Louis
St. Louis, MO 63130
[email protected]
Shibin Parameswaran
Department of Electrical and Computer Engineering
Unive
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
SARCASM DETECTOR
Project Documentation
Sarcasm Detection
Ishan Chutani
2013A7PS112P
1
SARCASM DETECTOR
ACKNOWLEDGEMENTS
I would like to thank Dr VS Nirban for me giving the opportunity to take up this
project. I hereby declare that I have done this projec
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
ISHAN CHUTANI
2013A7PS112P
1
Foundation and History of ISIL
(There is no god but Allah, Muhammad is the
messenger of god.)
ISIL has had many names in the past. It was formed by
Abu Musab alZarqawi, who was a Jordanian born
terrorist who had previously be
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
SARCASM DETECTOR
Project Documentation
Sarcasm Detection on Twitter
Discourse
Ishan Chutani
2013A7PS112P
1
SARCASM DETECTOR
ACKNOWLEDGEMENTS
I would like to thank Dr VS Nirban for me giving the opportunity to take up this
project. I hereby declare that I
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
ISHAN CHUTANI
2013A7PS112P
1
ISIL: ISLAMIC STATE OF IRAQ AND LEVANT
Aims and Hypothesis
Blessed Flagcovers all eastern and western extents
of the earth, filling the world with the truth and justice
of Islam and putting an end to the falsehood and
tyranny
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Chapter 5
Divide and Conquer
Divide and conquer refers to a class of algorithmic techniques in which one
breaks the input into several parts, solves the problem in each part recursively,
and then combines the solutions to these subproblems into an overall
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Chapter 6
Dynamic Programming
We began our study of algorithmic techniques with greedy algorithms, which
in some sense form the most natural approach to algorithm design. Faced with
a new computational problem, weve seen that its not hard to propose multi
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Chapter 4
Greedy Algorithms
In Wall Street, that iconic movie of the 1980s, Michael Douglas gets up in
front of a room full of stockholders and proclaims, Greed . . . is good. Greed
is right. Greed works. In this chapter, well be taking a much more unders
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
EfZZCV/M>\/XV:<M,LQ7GE/S 714,148,61T'g 61
M (794 MMM 6T ham 59 @ we
0%? t/Y6 Ome Maacen? /1m97)4. W
M31: 77% WW Defkamid 34%
G CU/E) QWW CerCu/EL) uolcblv'
<q,v>eE? i461 an? i7: a WM kak
outmw ha dag MW Mam b/QAGLWQQW
tmeio/l k' 59w, ~M 5"! 9W
17: 0&3A
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
W 9630g =mwmw~ g
Mam (9+ o\ AW ggfmw Ammm'le
m va won/Me cfw_ML Ia:</Loz/~'/V/Q>ZAW
m cg; PMWV! galabk? a an W07
g cfw_mg/M7 v14) Le/f V:/1&r/Y\j
 ' own ZI/L,~/< 9 Wm; '
W0 cfw_OW/L QLVQ/JLCQ j/J V/ mwwwamk
. ()4 99
may We, Malia Mcgmeo WWW
Egank5]
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
DQMEmwJ\(DFS= $FSstL&%A
W, 1%, WWWFWWI HRgFS:
Dam Ev :5 <1, 01>: WV ad v. 7r wvmj /
m ue/MA WW 5+A DPS W c~ '
1e MW WW? WA ee/M
QWVW@.WQcfw_%QZMEWNtQcfw_J .
LQ kw; $36194" 9farm :S'"('/1\e/HW fmzvtam
mac/J03 e We; 44M]: aAWOQ [amoQ Miami
Eff aw WWW/0 WW
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
W5
W M $30457 be ox
W mi MMM Wmx M, We, gm
Weagge (onyx/g, wamMW Moms)
edfa/w? Jote W'QWMJL Hm 2, Memoft;
QMQWO. aka/154d TEL52%; Waldo/bf 071th:
. @2967
3%WMQ/VMRAM W, LSM wand/(K
mm w W M "S/wm mm. was
(MW cfw_mom &WNW"MEW%TW
WWW'SMKM'M Uee. 1'
CM; W 9
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
R script to convert data from local disk into the format of the ML Studio of Azure:
The script was written in the template in ML Studio, I just changed the input parameters to read from our
file instead of the Azure file.
The Columns s1 till s28 refer to
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Module5 (MidTerm Problems)
Sec 1.1 Problem 2
3 d2x/dt2 +4dx/dt +9x = 2 cos(3t)
This is an ordinary differential equation with order =2 , highest derivative
Where t is the independent variable and x is the dependent variable and also it is a linear equat
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
November 6, 1999
1999 ACM Programming Contest
Problem 1
input file = prog1.in
Mid Atlantic Region
Where on Earth am I?
You and some friends are on a back country camping trip when your GPS unit fails. It turns out
that the onboard software was corrupted b
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
ACM International Collegiate
Programming Contest 1999/Y2K
Sponsored by IBM
Arab and African Regional Contest
Supported by Omnidata, RAM, BP and EMO
Al Akhawayn University in Ifrane, Morocco
November 21st, 1999
This problem set should contain 7 problems an
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Problems Statement: CERC 1999
ACM International Collegiate Programming Contest
Sponsored by IBM
Central Europe 1999 Regional Contest
Czech Technical University in Prague
Dept. of Computer Science, Faculty of Electrical Engineering
November 13, 1999
Contes
Birla Institute of Technology & Science, Pilani  Dubai
Data Structure & algorithm
CSC F211

Winter 2016
Shabbir Husain's Report
ACM Regional Asia 99
Dhaka Site
Problem A
Problem B
Problem C
Problem D
Problem E
Problem F
http:/www.northsouth.edu/Problems.htm [2/23/2000 3:56:02 PM]
From NonTeX to TeX
ACM International Collegiate Programming Contest, 1999
Asia