Machine Learning

John Samuel
CPE Lyon

Year: 2018-2019
Email: john(dot)samuel(at)cpe(dot)fr

Artifical Intelligence

Knowledge Representation

• A machine-readable form of representing knowledge of world or a domain.
• Example: semantic nets, ontologies
• Trade-off between expressivity and practicality

Natural Language Processing

• Analysing and understanding natural (human) language
• Human-machine interaction
• Syntax of a language
  • Parsing
  • Part of speech tagging
• Semantics of a language
  • Machine translation
  • Named entity recognition
  • Sentiment analysis

Named Entity Recognition

Extracting named entities and assigning them to specific categories.

Recommendation Engines

• Primary goal: reducing information overload by providing filtered and relevant information
• Predicts user's preference
• Approaches
  • Collaborative filtering: based on multiple user ratings
  • Content-based filtering: based on user profiles

Rules engine

• Consisting of rules and constraints
• Verifies that at any given instant, the system is consistent

Logic Programming

• Propositional logic
• First-order logic (FOL, predicate logic)

Prolog

• Declarative programming language
• Logic programming language: based on first-order logic
• Developed in 1972 by Alain Colmerauer
• Expressed in terms of relations: facts and rules
• Used for theorem proving, expert systems and NLP

Prolog: Data types

• Atom
• Numbers
• Variables
• Compound term (e.g., strings, lists)

Prolog: Rules

A Prolog program contains clauses of the following form.

Head :- Body.

• Body may contain one or more predicates using conjunction and disjunction
• Head does not contain any conjunction and disjunction

Prolog: Facts

A clause with empty body is called fact.

cat(bob).
cat(alice).


Prolog: GNU Prolog

$ prolog
GNU Prolog 1.4.5 (64 bits)
Compiled Feb 5 2017, 10:30:08 with gcc
By Daniel Diaz
Copyright (C) 1999-2016 Daniel Diaz
| ?- [user].
compiling user for byte code...
cat(tom).
cat(alice).

user compiled, 2 lines read - 241 bytes written, 12239 ms

(4 ms) yes
| ?-



Prolog: GNU Prolog

?- cat(X).

X = tom ?

yes
| ?- cat(bob).

no


Prolog: GNU Prolog

| ?- [user].
compiling user for byte code...
cat(tom).
cat(alice).
allcats(L) :- findall(X, cat(X), L).

user compiled, 3 lines read - 490 bytes written, 10638 ms

yes
| ?- allcats(L).

L = [tom,alice]

yes


Prolog: GNU Prolog

| ?- [user].
compiling user for byte code...
friend(bob, alice).
friend(alice, kevin).
friend(bob, thomas).
friend(bob, peter).

user compiled, 4 lines read - 486 bytes written, 77256 ms

(10 ms) yes
| ?- friend(bob, X).

X = alice ? a

X = thomas

X = peter

(1 ms) yes


Web

• https://en.wikipedia.org/wiki/Recommender_system
• https://en.wikipedia.org/wiki/Artificial_intelligence
• https://en.wikipedia.org/wiki/Logic_programming
• https://en.wikipedia.org/wiki/Knowledge_representation
• https://en.wikipedia.org/wiki/Prolog

Colors

• Color Tool - Material Design

Images

• Wikimedia Commons