Disclaimer :

1
Disclaimer The jokes during the seminar were
generated either by AI (Artificial
Intelligence) or by AI (Aadityas
Intelligence). The bottomline, AI is good.
2
Humour AI
Devshree D Sane 08305059 devshreedsane_at_cse
Aditya M Joshi 08305908 adityaj_at_cse
Under the guidance of Dr. Pushpak Bhattacharyya
3
Motivation

• Trust
• Interpersonal Attraction
• Stress Release

Why Humour? Why AI? Why Humour AI?

• Use Existing Intelligent
• Systems - humans
• Model Intelligent systems
• as close as possible to them

• Computers As Social Actors
• Cognitive science studies

Humour is a powerful weapon - you can even break
ice with it.
4
Scope of the Seminar
Humour Generation
Humour Recognition
5
Humour AI
Humour theory
Computational Humour
JAPE-1
HAHAcronym
Humour theory
Humour Recognition
Applications of Computational humour
6
What is humour?

• Wit
• Mirth
• Laughter
• Manner

Components Humour Research Challenges

• Humour theory
• Sociological Research
• Gelotology (Health effects)
• Computational humour

• Different to different people
• Different at different times

7
Theories of humour

• Focus on feelings necessary
• for humour.
• Mixture of pleasure and pain
• at the base of amusement

Superiority theory Relief theory Incongrui
ty theory

• Focus on effect of humour
• Release of nervous energy

• Gives a necessary condition
• For humour a twist.
• Humour arises from showing
• something absurd from
• something that is not.
• Based on contradiction of some
• sort.

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.)
8
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. ?
9
Humour AI
Humour theory
Computational Humour
JAPE-1
HAHAcronym
Computational Humour
Humour Recognition
Applications of Computational humour
10
What is computational humour?

• Using computers in humour
• research.
• Modelling humour in a
• computationally tractable way.

Definition Areas Our Focus

• Humour Generation
• Humour Recognition

• Out of all forms,
• text-based / Verbal Humour
• Humour in one-liners

11
Computational Humour Linguistic Ambiguity
A word is ambiguous if it has more than one
meaning. (Ambiguous is thankfully not
ambiguous. ? )

• Same sounds, different
• meaning.
• Three ways
• Syllable substitution
• E.g. What do short-sighted ghosts wear?
• Spooktacles.
• Word substitution
• E.g. How do u make gold soup?
• Put 14 carrots in it.
• Metathesis (Reversal of
• sounds)

Phonological Morphological Syntactic

• Words with same surface
• structure.
• E.g. The book is read / red.

• As a result of structure or
• syntax of sentence.
• Example Squad helps dog
• bite victims.

12
Humour AI
Humour theory
Computational Humour
JAPE-1
HAHAcronym
JAPE-1
Humour Recognition
Applications of Computational humour
13
JAPE-1

• Generates question-answer style puns using
  phonological similarities

• For example,
• What do you give an elephant thats exhausted?
• Trunkquillizers.

14
JAPE-1 Units

• A set of lexemes.
• Lexeme is an abstract entity,
• roughly corresponding to a
• meaning or a phrase.
• In addition, a homonym base.

Lexicon Schemata Template
A set of relationships which must hold between
the lexemes
To produce the surface form of a joke from the
lexemes and relationships specified in an
instantiated schema.
15
JAPE-1 Example
Lexeme jumper Synonym Sweater Category
Noun Countable Yes Specifying adjective Warm
Lexicon Schemata Template
What will you get if you cross ____ and
____? Answer _______
16
Humour AI
Humour theory
Computational Humour
JAPE-1
HAHAcronym
HAHAcronym
Humour Recognition
Applications of Computational humour
17
HAHAcronym

• European project
• Humorous Agent for
• Humourous Acronyms.
• Acronym Ironic
• Re-analyzer and generator

About HAHAcronym Features Examples

• Makes fun of existing
• acronyms.
• Produces new acronyms
• based on concepts provided
• by the user.

• ACM
• We say Association for
• Computing machinery
• HAHA says Association for
• Confusing machinery

http//www.haha.itc.it
18
HAHAcronym Concepts

• group of data elements
• that are considered semantically
• equivalent for the purposes of
• information retrieval.
• Eg. Person, Human, Individual

Synset WordNet WordNet Domains

• A large database of English.
• Words are grouped into sets
• of synonyms (synsets),
• each expressing a distinct concept.
• Synsets are interlinked by meansof semantic and
  lexical relations.

• Augment WordNet with
• domain labels.
• Example, the word bank has
• two labels
• Economy and Geology.

19
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. 2.
Reproducing rhyme and rhythm. 3. Adjectives
antonym clustering and semantic relations in
WordNet.
Recognizes individual constituents such as NP,
VP, etc. using acronym grammar.
20
HAHAcronym Examples of Acr. Modification

• Close Combat Tactical Trainer
• Close Combat Theological Trainer
• Two changes antonym strategy
• for first adjective and semantic
• opposition found in religion' domain
• for tactical to theological.

CCTT CHI
Computer Human Interface. Computer
Harry_Truman Interface. Unexpected result
due to rhyming of "human" to "harry_truman"
21
HAHAcronym Acronym generation
Input Main concept Attribute Output A new
funny acronym.
Input/Output Example
Concept Writing Attribute Creative CAUSTIC
Creative Activity for Unconvincingly
Sporadically Talkative Individualistic Commercials
22
Humour AI
Humour theory
Computational Humour
JAPE-1
HAHAcronym
Humour Recognition
Applications of Computational humour
Humour Recognition
23
Humour Recognition

• Humorous data One-liners
• Non-humorous data
• Sentences from BNC
• Sentences from Reuters
• Proverbs

Experiment Strategies

• Stylistic features
• Content-based features
• Combination of the two

24
Humour Recognition Strategies

• Antonymy
• ( A clean desk is a sign of
• cluttered drawer.)
• Alliteration
• ( Infants dont enjoy infancy
• as much as adults do adultery.)

Stylistic Features Content-based features

• Naïve Bayes
• Uses joint probability of
• words and documents.

25
Humour Recognition Results
26
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.

27
Humour AI
Humour theory
Computational Humour
JAPE-1
HAHAcronym
Humour Recognition
Applications of Computational humour
Applications of Computational humour
28
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

29
Applications ALICE

• Human Its a funny thing!
• ALICE I thought it was too.
• Human Then why didnt 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.

30
Applications ALICE

• Human Will you marry me?
• ALICE Why dont you just download me?

31
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.
32
We learnt

• The past was .
• The future is .

computerization
humanization
33
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 (AAAI-94),
  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

34
Humour AI
Questions? Comments? Suggestions?
The past was computerization. The future is
humanization. ?