Transcript Humour - CSE, IIT Bombay

Disclaimer :
The jokes during the seminar
were generated either by AI (Artificial Intelligence)
or by AI (Aaditya’s Intelligence).
The bottomline, AI is good.
Humour & AI
Aditya M Joshi
Devshree D Sane
08305908
08305059
adityaj@cse
devshreedsane@cse
Under the guidance of
Dr. Pushpak Bhattacharyya
Motivation
Why Humour?
• Trust
•• Interpersonal
Attraction
Use Existing Intelligent
Systems - humans
Why Humour & AI? • Stress Release
• Computers As Social Actors
• Model Intelligent systems
• Cognitive science studies
as close as possible to them
Why AI?
“Humour is a powerful weapon - you can even break ice with it.”
Scope of the Seminar
Humour
Recognition
Humour
Generation
Humour & AI
Humour theory
Computational
Humour
Humour
Recognition
JAPE-1
Applications of
Computational
humour
HAHAcronym
What is humour?
Components
• Wit
Humour Research •
•
Challenges
•
•
•
•
•
Mirth
Humour theory
Sociological
Research
Laughter
Gelotology (Health effects)
Different to different people
Computational
humour
Manner
Different at different times
Theories of humour
Superiority theory
Relief theory
Incongruity theory
Dry humour is a form of
humour which is narrated
as if it is not a joke at all
(i.e. narrated in a serious
tone, perhaps.)
•Focus on feelings necessary
•forGives
aonnecessary
condition
Focus
effect of humour
humour.
For
humourof
– nervous
a ‘twist’. energy
• Release
•• Humour
arises
fromand
showing
Mixture of
pleasure
pain
something
from
at the baseabsurd
of amusement
something that is not.
• Based on contradiction of some
sort.
Examples of jokes
Incongruity theory: "Some people can tell what time it is by looking at
the sun. But I have never been able to make out the numbers."
Superiority theory: All the “blonde” or “Sardarji” jokes that are cracked.
Relief theory: The “battle-of-the-sexes” jokes
A pun in Hindi:
Sawaal: Shahrukh Khan ne ek sansthaa ko Rs.10000 ka chandaa diya.
Us chande ko kya kehte hain?
Jawab: “KHAN-DAAN”. 
Humour & AI
Humour theory
Computational
Humour
Humour
Recognition
JAPE-1
Applications of
Computational
humour
HAHAcronym
What is *computational* humour?
Definition
Areas
Our Focus
• Using computers in humour
research.
•• Modelling
Humour Generation
humour in a
computationally
tractable way.
• Out of all forms,
• Humour Recognition
text-based / Verbal Humour
• Humour in one-liners
Computational Humour – Linguistic Ambiguity
A word is ambiguous if it has more than one meaning. (‘Ambiguous’ is
thankfully not ambiguous.  )
Phonological
Morphological
Syntactic
• Same sounds, different
meaning.
• Three ways:
•Words with same surface
•Syllable substitution
structure.
E.g. What do short-sighted ghosts wear?
• AsSpooktacles.
a result of structure or
E.g.
: The
book
is read / red.
syntax
ofsubstitution
sentence.
•Word
E.g. How do“Squad
u make gold
soup?
• Example:
helps
dog
14 carrots in it.
bitePut
victims.”
•Metathesis (Reversal of
sounds)
Humour & AI
Humour theory
Computational
Humour
Humour
Recognition
JAPE-1
Applications of
Computational
humour
HAHAcronym
JAPE-1
• Generates question-answer style puns using
phonological similarities
•For example,
What do you give an elephant that’s exhausted?
Trunkquillizers.
JAPE-1 : Units
Lexicon
•A set of lexemes.
Schemata
•Lexeme is an abstract entity,
roughly corresponding to a
meaning
a phrase. which
A set ofor
relationships
Template
must
hold
between
lexemes
•In addition,
aproduce
homonym
base.
To
thethe
surface
form of a joke from the lexemes
and relationships specified
in an instantiated schema.
JAPE-1: Example
Lexicon
Schemata
Template
Lexeme : jumper
Synonym : Sweater
Category: Noun
Countable:
“What will Yes
you get if you cross
Specifying
adjective
: Warm
____
and ____?”
Answer: _______
Humour & AI
Humour theory
Computational
Humour
Humour
Recognition
JAPE-1
Applications of
Computational
humour
HAHAcronym
HAHAcronym
About HAHAcronym
Features
Examples
http://www.haha.itc.it
• European project
• Makes fun of existing
• Humorous Agent for
acronyms.
Humourous Acronyms.
ACM
:
• Acronym
Ironic
• Produces new acronyms
•Re-analyzer
We say: Association
for
and generator
based on concepts provided
Computing machinery
by the user.
• HAHA says: Association for
Confusing machinery
HAHAcronym : Concepts
Synset
• group of data elements
that are considered semantically
WordNet
•A
large database
English.of
equivalent
for the of
purposes
information retrieval.
WordNet Domains •Augment
with
•Words
areWordNet
grouped
into
sets
• Eg. Person,
Human,
Individual
domain
labels.
of synonyms
(synsets),
each expressing a distinct concept.
•Example, the word ‘bank’ has
two
labelsare
– interlinked by means
•Synsets
Economy
and
Geology.
of semantic
and
lexical relations.
HAHAcronym : Acronym modification
Acronym parsing
and
construction of
logical form
Choice of what
to modify
and what to keep
unchanged
Substitutions
1. Using semantic field oppositions.
Recognizes individual constituents such as NP, VP, etc.
2. Reproducing rhyme and rhythm.
using acronym grammar.
3. Adjectives: antonym clustering and semantic relations
in WordNet.
HAHAcronym : Examples of Acr. Modification
CCTT
CHI
•Close Combat Tactical Trainer
•Close
CombatHuman
Theological
Trainer
Computer
Interface.
Computer Harry_Truman Interface.
Two changes: antonym strategy
for first adjective and semantic
Unexpected
result:
opposition
found in ‘religion'
domain
duefortotactical
rhyming
of "human"
to theological.
to "harry_truman"
HAHAcronym : Acronym generation
Input/Output
Example
Input: Main concept
+ Writing
Concept:
Attribute
Attribute:
Creative
Output: A–new
funny acronym
CAUSTIC
Creative
Activity. for
Unconvincingly Sporadically
Talkative Individualistic
Commercials
Humour & AI
Humour theory
Computational
Humour
Humour
Humour
Recognition
JAPE-1
Applications of
Computational
humour
HAHAcronym
Humour Recognition
Experiment
Strategies
•
•
•
Stylistic features
Humorous data : One-liners
Non-humorous
:
Content-based data
features
• Sentences from BNC
• Sentences from
• Combination
of theReuters
two
• Proverbs
Humour Recognition : Strategies
Stylistic Features
• Antonymy
Content-based features ( A clean desk is a sign of
cluttered drawer.)
• Naïve Bayes
• Uses joint probability of
• Alliteration
words and documents.
( Infants don’t enjoy infancy
as much as adults do adultery.)
Humour Recognition : Results
Humour Recognition : N-grams
Naïve Bayes assumes independent words.
N-gram :
• Allows a computer to process ‘word sequences’.
• Uses conditional probability to predict word based on N-1
previous words.
• Allows processing of ‘multi-words’.
• Example: wild card, freak out, etc.
Humour & AI
Humour theory
Computational
Humour
Humour
Recognition
JAPE-1
Applications
Applications of
of
Computational
humour
HAHAcronym
Applications
• Advertisements
• Human-Computer Interaction
– Text-based Interface
– Embodied Agents (Helpdesks)
– Chat bots (A.L.I.C.E.)
(Artificial Linguistic Internet Computer entity)
PLEASE visit: www.alicebot.org
Applications : ALICE
Human: It’s a funny thing!
ALICE: I thought it was too.
Human: Then why didn’t you laugh?
ALICE: Perhaps it is not part of my
programming.
Human: So you are not programmed to
laugh at all?
ALICE: Interesting deduction. Ha Ha.
Applications : ALICE
Human: Will you marry me?
ALICE: Why don’t you just download me?
Conclusion
• Humour is a combination of intellect and emotions. It is
subject to time, individual experience and perception.
• Humour generation produces instances of humour.
Textual humour is generated in systems like JAPE and
HAHA.
• Humour recognition takes help of machine learning
techniques to understand the ‘humour’ content of a
situation/statement.
A conclusion is simply the place where you got tired of thinking.
We learnt…
The past was ‘computerization ’.
The future is ‘ humanization ’.
References
Humour Theory and Computational Humour:
www.dcs.gla.ac.uk/~kimb/dai_version/dai_version.html
JAPE-1:
• Kim Binsted and Graeme Ritchie. An implemented model of punning
riddles. In Twelfth National Conference on Artificial Intelligence (AAAI94), pages 1-6, 1994.
HAHAcronym:
• An Experiment in Automated Humorous Output Production. Oliviero
Stock and Carlo Strapparava. In IUT 2003, pages 1-3, 2003.
Humour Recognition:
• Making Computers Laugh. Rada Mihalcea and Carlo Strapparava. In
Proceedings of HLT/EMNLP, pages 531-538, 2005.
• www.wikipedia.org
Humour & AI
Questions?
Comments?
Suggestions?
The past was computerization. The future is humanization. 