Itv0060 2009 arhiiv
Link: http://www.lambda.ee/index.php/Teadmiste_otsing,_formaliseerimine_ja_hoidmine or http://www.lambda.ee/index/itv0060
Alternative names and codes:
- Tehisintellekti algoritmid ITT0020
- Intelligent Systems ITX8075
NB! Siin on 2009 aasta arhiiv, mitte uus ja kehtiv info!
Nimi Matrikk Prax Eksam Kokku Hinne SJ 40602 40 75 85 4 RP 50606 40 78 86,8 4 MM 50947 35 65 74 3 AK 90206 20 20 32 0 AH 20576 20 50 50 1 RA 990746 40 90 94 5 SK 81897 40 40 64 2 DT 81928 35 42 60,2 2 TO 30633 40 90 94 5 RL 930711 40 60 76 3 KO 40980 40 30 58 1 MA 20446 40 82 89,2 4 KP 40645 40 67 80,2 4 DK 41008 40 40 64 2 AS 81889 35 80 83 4 AN 40 75 85 4 AV 50668 40 90 94 5 JS 50658 40 82 89,2 4 JV 81900 30 70 72 3
Materials for exam
Use and read the following
Philosophy around AI
Just for background, no questions during the exam:
RDF, RDFS, OWL and conversion to 1st order logic
Understand RDF and RDFS:
Understand basics of owl:
RDFa, reading and using RDF data from web pages
Understand main parts:
- wiki intro
- Media:portaalidekoosvoime.ppt or as pdf Media:portaalidekoosvoime.pdf
- w3c rdfa primer
- microformats: a similar technology
RDF relation to logic
Understand a bit of:
Semantics of ordinary relational base
First order logic in query engines and planning
- Geoff reasoning course notes
- planning and blocks world
- blocks world axiom/query examples from lecture
Different languages for 1st order logic: KIF, CL, restricted english jms
Representing knowledge and belief
Understand basics of discrimination tree trie-type term indexes from
Game playing search (only for ITX8075)
Learning (only for ITX8075)
Understand basics from detailed intros:
Time, place, result
Lectures: Every friday 8.00, room IT-140
Practical work: every second (even weeks) Friday 10.00 room IT-213D
Practical work will give 40% and exam 60% of points underlying the final grade. NB! Course ITV0060 and ITT0020 attendees may optionally do the third lab and get 60% from labs and 40% from exam.
- For ITV0060 and ITT0020: two labs on the same theme.
- For ITX8075: same two labs plus a third lab below.
- As noted above, course ITV0060 and ITT0020 attendees may optionally do the third lab and get 60% from labs and 40% from exam.
Date: 13 March.
Implement a system which can read html annotated according to the RDFa standard. The data read should be saved and printed in several formats:
- rdf xml syntax
You will have to either find suitable annotated html pages or annotate some existing pages yourself (at least three pages).
- http://www.w3.org/TR/rdf-primer/ RDF tutorial first parts
- http://en.wikipedia.org/wiki/RDFa wikipedia article on RDFa
- http://www.youtube.com/watch?v=ldl0m-5zLz4 and http://www.youtube.com/watch?v=mxE3FeOyS-E video tutorials of rdfa
- Media:portaalidekoosvoime.ppt või pdf kujul Media:portaalidekoosvoime.pdf (Estonian)
You may also look at some practical non-obligatory programming suggestions in Estonian: rdfa esimese praksi soovitusi.
Using the results of first lab for rule-based derivation of new information.
This lab is obligatory only for the ITX8075 students obtaining 5 AP from the course! You can do it alone or together with another student: pairs are OK, three or more persons not. Both students must be present while presenting the result.
The lab depends on your background. Concretely, choose the first case which applies to you:
- You are studying in the IT department and have never programmed a turn-based two-player thinking game. In this case you should program the five-in-a-row game as explained in Praktikumitöö: mõtlemismäng 2005. This option is also allowed - but not obligatory - for non-IT students who have never programmed a two-player thinking game before.
- You are studying in the IT department and have programmed a turn-based two-player thinking game. In this case you should experiment with the Attempto system in the following way:
- Create a set of English sentences which covers the meaning of your rules from the second lab as closey as possible.
- Use the Attempto system to transform these sentences into a formal rule system.
- Write a program which would transform the sentences from the Attempto-given rule system into the form acceptable by Otter or Prover9.
- All the other cases: you will have to do a project from the similar MIT course. Download the MIT course materials and choose one of the following projects:
Intro and paradigms of knowledge representation
Overview of knowledge representation: http://en.wikipedia.org/wiki/Knowledge_representation
- relational, network, object databases
- semantic networks
- frame systems
- property/value pairs
- rule/expert systems
- rdfs and owl
Core search, search in game trees
Lecture notes: http://www.lambda.ee/index.php/Pilt:Search.ppt
- wikipedia article on search algorithms
- wikipedia article on the A* algorithm
- Norvig AIMA constraint satisfaction chapter
- Minimax & alpha-beta: minimax wikipedias, a-b in wikipedia, applet demo, a-b ok explanation with
- Chinook: http://www.cs.ualberta.ca/~chinook/
- Ph.D. thesis about game programming: http://fragrieu.free.fr/SearchingForSolutions.pdf
- propositional, first order, higher order
- proofs and deriving new information
- automated reasoning
- relational data and rdf in first order logic
logic- and reasoning-based methods
Different property/value pair representations
Methods and their relations
- association lists
- google base
Specific versions of 1st order logic and rule systems
- restricted english systems
Details of RDF, RDFS, RDFa
Representing time, probability, epistemics, context
Search and indexing