My PhD Study - Revision history

Slavkoz: Slavkoz moved page PhD Study to My PhD Study without leaving a redirect

2022-08-02T20:36:59Z

Slavkoz moved page PhD Study to My PhD Study without leaving a redirect

← Older revision		Revision as of 22:36, 2 August 2022
(No difference)

Slavkoz at 20:32, 2 August 2022

2022-08-02T20:32:31Z

← Older revision		Revision as of 22:32, 2 August 2022
Line 1:		Line 1:
	[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]		[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]

	My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [{{~~filpath~~:PhD_thesis_slavko.pdf}} PhD thesis is available online] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].		My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [{{filepath:PhD_thesis_slavko.pdf}} PhD thesis is available online] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].

	Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.		Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

Slavkoz at 20:31, 2 August 2022

2022-08-02T20:31:59Z

← Older revision		Revision as of 22:31, 2 August 2022
Line 1:		Line 1:
	[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]		[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]

	My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [~~File~~:PhD_thesis_slavko.pdf PhD thesis is available online] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].		My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [{{filpath:PhD_thesis_slavko.pdf}} PhD thesis is available online] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].

	Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.		Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

Slavkoz at 20:31, 2 August 2022

2022-08-02T20:31:21Z

← Older revision		Revision as of 22:31, 2 August 2022
Line 1:		Line 1:
	[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]		[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]

	My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [[File:PhD_thesis_slavko.pdf\|PhD thesis is available online]] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].		My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [File:PhD_thesis_slavko.pdf PhD thesis is available online] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].

	Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.		Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

Slavkoz at 20:28, 2 August 2022

2022-08-02T20:28:50Z

← Older revision		Revision as of 22:28, 2 August 2022
Line 1:		Line 1:
	[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]		[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]

	My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [[~~PhD thesis slavko~~.pdf\|PhD thesis is available online]] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].		My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [[File:PhD_thesis_slavko.pdf\|PhD thesis is available online]] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].

	Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.		Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

Slavkoz at 20:28, 2 August 2022

2022-08-02T20:28:09Z

← Older revision		Revision as of 22:28, 2 August 2022
Line 1:		Line 1:
	[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]		[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]

	My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My PhD thesis is available online.		My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My [[PhD thesis slavko.pdf\|PhD thesis is available online]] along with [https://www.youtube.com/watch?v=3R3Ob5tHL7c the public defense].

	Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.		Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

Slavkoz at 20:25, 2 August 2022

2022-08-02T20:25:22Z

← Older revision		Revision as of 22:25, 2 August 2022
Line 1:		Line 1:
	[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]		[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]
	~~I am a postdoctorate researcher at the University of Ljubljana, Faculty of Computer and Information Science.~~ My research ~~interests are~~ related to ~~broad Information Retrieval and Information Extraction fields, which I also researched in my [http://eprints.fri.uni~~-lj.~~si/1160/ diploma thesis]~~ and ~~[http://eprints~~.~~fri~~.~~uni-lj.si/2889/~~ PhD thesis].
			My PhD research was related to an iterative ontology-based information extraction. I was working on named entity recognition, relationship extraction and coreference resolution. After developing the novelties I merged the methods into and iterative system. My PhD thesis is available online.

	Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.		Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

Slavkoz at 20:21, 2 August 2022

2022-08-02T20:21:28Z

← Older revision		Revision as of 22:21, 2 August 2022
Line 1:		Line 1:
			[[Image:Shutterstock_128869777_paidImage_small.jpeg\|thumb\|''Data is the new oil.'' — Clive Humby, Kellogg School, 2006.\|300px\|right]]
			I am a postdoctorate researcher at the University of Ljubljana, Faculty of Computer and Information Science. My research interests are related to broad Information Retrieval and Information Extraction fields, which I also researched in my [http://eprints.fri.uni-lj.si/1160/ diploma thesis] and [http://eprints.fri.uni-lj.si/2889/ PhD thesis].

			Information Extraction (IE) refers to automatic extraction of structured information from unstructured sources. As a task it can also be seen as flling slots into a database from text. It must pre-process, recognize and convert information from textual documents (e.g web pages, reports, books), structural (e.g. web page structure, indexes) or usage data (e.g. query logs) into human and machine understandable format. As a family of techniques IE combines segmentation, classification, association and clustering. They can be roughly divided into pattern-based and machine learning-based (ML) approaches. The first use manually defined rules or can also learn them for specific type of documents using seed expansion. The latter consist of probabilistic (e.g. sequence models) and induction (e.g. linguistic, structural models) approaches and are currently the main focus of the research in IE community. In knowledge management and semantic web, a machine can ''understand'' the data if it is represented as an ontology. Therefore IE techniques can be used for automatic ontology creation and also population.

	== 1st year (2010/2011) ==		== 1st year (2010/2011) ==

Slavkoz: Created page with "== 1st year (2010/2011) == === Veščine v znanstvenem delu 1 === * Osnovni principi znanstvenega sporočanja (prof. Jurišić) [{{filepath:01_DN_SpremljanjeRaziskovalnePeri..."

2021-11-15T16:54:36Z

Created page with "== 1st year (2010/2011) == === Veščine v znanstvenem delu 1 === * Osnovni principi znanstvenega sporočanja (prof. Jurišić) [{{filepath:01_DN_SpremljanjeRaziskovalnePeri..."

New page

== 1st year (2010/2011) ==

=== Veščine v znanstvenem delu 1 ===

* Osnovni principi znanstvenega sporočanja (prof. Jurišić) [{{filepath:01_DN_SpremljanjeRaziskovalnePeriodike_SlavkoZitnik.pdf}} Naloga]
* Sporočanje v angleškem jeziku, dobra praksa in tipične napake (viš. pred. Štros Bračko) [{{filepath:02_DN_anglescina_SlavkoZitnik_optimized.pdf}} Naloga]
* Javne bibliografske baze in njihova uporaba v znanstveno raziskovalnem delu, osnove scientometrije (doc. Demšar) [{{filepath:03_DN_IndeksiranostRevij_SlavkoZitnik.pdf}} Naloga]
* Ustno sporočanje, elementi dobre ustne predstavitve oz. predavanja (doc. Demšar) [{{filepath:}} Naloga]
* Orodja za urejanje znanstvenih besedil, orodja za vodenje in urejanje bibliografskih zapisov (izr. prof. Lotrič) [{{filepath:04_DN_UrejanjeBesedil_SlavkoZitnik.zip}} Naloga]
* Osnovni principi pisnega znanstvenega sporočanja v znanosti in citiranje (prof. Jurišić) [{{filepath:05_DN_clanki_SlavkoZitnik.pdf}} Naloga]
* Sporočanje v obliki plakata 1,2 (doc. Lebar Bajec) [{{filepath:Priloga.Slavko.Zitnik.pdf}} Priloga] [{{filepath:Poster.Slavko.Zitnik_small.pdf}} Plakat]
* Načrtovanje in pisanje doktorskih disertacij (prof. Solina) [{{filepath:07_DN_doktorati_SlavkoZitnik.pdf}} Naloga]
* Postopek recenziranja (prof. Leonardis) [{{filepath:08_DN_recenzija_SlavkoZitnik.pdf}} Naloga]
* Modeli financiranja raziskav doma in v svetu (prof. Leonardis)
* Etika v znanosti in raziskovanju (prof. Jurišić) [{{filepath:ZitnikSlavkoDNetika.pdf}} Naloga]
* Znanost in mediji (prof. Solina)

=== Seminar 1 ===

V okviru predmeta boste pripravili pregledno predavanje za izbrano področje računalništva. Predavanje bosta skupaj predstavila dva študenta, ki najprej izbereta področje (temo) in predvideni naslov predstavitve, poiščeta vsaj tri pregledne revijske znanstvene članke z izbranega področja, skupaj pripravita predstavitev in to skupaj predstavita pred avditorijem. Predstavitev naj bo dolga 20 minut (+5 minut za vprašanja in komentarje).

Udeležba na vseh seminarjih je za vse študente III. stopnje obvezna. Za pozitivno opravljen predmet je potrebna:

* pravočasna prijava predstavitve (1. korak),
* naložitev predstavitve v PDF obliki vsaj 72 ur pred predstavitvijo (2. korak),
* ustna predstavitev pred avditorijem (3. korak),
* prisotnost na vseh predstavitvah v okviru Seminarja 1, oziroma, če ste na kakšni predstavitvi manjkali, prebran ustrezni pregledni članek in oddan povzetek članka.

Literatura:
* Bourne PE (2007) Ten Simple Rules for Making Good Oral Presentations, PLoS Computational Biology 3(4): e77.

[{{filepath:BlagusZitnik-clanek.pdf}} Članek]
[{{filepath:BlagusZitnik.pdf}} Predstavitev (pdf)]
[http://prezi.com/67ymzzpklzp4/analiza-kompleksnih-omrezij-s-primeri-uporabe/ Prezi link]

=== Veščine v znanstvenem delu 2 ===

Pisanje prijave na ARRS projekt.

[{{filepath:Vescine2_FazaI_slavko.doc}} Prva faza]
[{{filepath:Vescine2_FazaII_slavko_popravljeno.doc}} Druga faza]

=== Seminar 2 ===

Na Seminarju 2 bomo spisali članek, ga oddali v recenziranje, recenzirali, pripravili predstavitev in se udeležili (neformalne) mini konference. Vse bomo pripravili v angleščini. Tokrat bo vsak študent pripravil svoj članek in imel svojo predstavitev. Na članku in na predstavitvi naj bo študent edini avtor, sodelovanje mentorjev in morebitne ostale pomoči naj omenite v zahvali. Formalne objave člankov ne bo, zato lahko gradivo, ki ga boste pripravili za ta predmet sami ali pa skupaj s sodelavci uporabiti za pripravo članka, ki ga pošljete na neko mednarodno konferenco ali v revijo.

[{{filepath:ClanekSeminar2.pdf}} Članek]
[{{filepath:PreziSeminar2.pdf‎}} Predstavitev]
[http://prezi.com/1pa_4tedp-fn/intelligent-ontology-based-information-extraction/ Prezi predstavitev]

=== Pregledne teme ===
Izpit iz Preglednih tem računalništva in informatike preverja študentovo znanje s področij, ki jih pokriva tipični dodiplomski kurikulum računalništva in informatike. Študent na začetku prvega semestra reši kvize, s katerimi preveri svoje znanje s posameznih področij. Če se izkaže, da je njegovo znanje pomanjkljivo, nadoknadi zaostanek s študijem ustrezne literature ozitoma, po potrebi, konzultacijami z izvajalci predmeta.

Izpit je razdeljen na štiri sklope, sklopi na podsklope, ki predstavljajo različen delež sklopa. Za uspešno opravljen izpit je potrebno pozitivno opraviti vse štiri sklope.

[{{filepath:PregledneTeme.zip}} Testna vprašanja]

=== Matematične metode v računalništvu ===

[{{filepath:Naloga1.pdf}} Naloga 1 navodila] [{{filepath:01_DN_MMR_SlavkoZitnik.pdf}} Naloga 1 rešitev]

[{{filepath:Naloga2.pdf}} Naloga 2 navodila] [{{filepath:02_DN_MMR_SlavkoZitnik.pdf}} Naloga 2 rešitev]

[{{filepath:Naloga3.pdf}} Naloga 3 navodila] [{{filepath:03_DN_MMR_SlavkoZitnik.pdf}} Naloga 3 rešitev]

[{{filepath:Naloga4.pdf}} Naloga 4 navodila] [{{filepath:04_DN_MMR_SlavkoZitnik.pdf}} Naloga 4 rešitev]

[{{filepath:Naloga5.pdf}} Naloga 5 navodila] [{{filepath:05_DN_MMR_SlavkoZitnik.pdf}} Naloga 5 rešitev]

[{{filepath:Naloga6.pdf}} Naloga 6 navodila] [{{filepath:06_DN_MMR_SlavkoZitnik.pdf}} Naloga 6 rešitev]

[{{filepath:Naloga7.pdf}} Naloga 7 navodila] [{{filepath:07_DN_MMR_SlavkoZitnik.pdf}} Naloga 7 rešitev]

[{{filepath:Naloga8.pdf}} Naloga 8 navodila] [{{filepath:08_DN_MMR_SlavkoZitnik.pdf}} Naloga 8 rešitev]

[{{filepath:Naloga9.pdf}} Naloga 9 navodila] [{{filepath:09_DN_MMR_SlavkoZitnik.pdf}} Naloga 9 rešitev]

[{{filepath:Naloga10.pdf}} Naloga 10 navodila] [{{filepath:10_DN_MMR_SlavkoZitnik.pdf}} Naloga 10 rešitev]

[{{filepath:Naloga11.pdf}} Naloga 11 navodila] [{{filepath:11_DN_MMR_SlavkoZitnik.pdf}} Naloga 11 rešitev]

[{{filepath:Naloga12.pdf}} Naloga 12 navodila] [{{filepath:12_DN_MMR_SlavkoZitnik.pdf}} Naloga 12 rešitev]

[{{filepath:Naloga13.pdf}} Naloga 13 navodila] [{{filepath:13_DN_MMR_SlavkoZitnik.pdf}} Naloga 13 rešitev]

Seminarska naloga:
[{{filepath:Projekt_mmr_SlavkoZitnik.pdf}} Iskanje hamiltonovega cikla na ravninskih grafih z malo notranjimi točkami]
Izvorna koda se nahaja na: [https://github.com/szitnik/Hamiltonian-Cycle-with-Few-Inner-Points https://github.com/szitnik/Hamiltonian-Cycle-with-Few-Inner-Points]

=== Velika omrežja ===

Gradivo in vse dostopno na pajek.imfm.si.

[{{filepath:01_dn_VelikaOmrezja_SlavkoZitnik_v2.pdf}} Seminarska 1]
[{{filepath: 02_dn_VelikaOmrezja_SlavkoZitnik.pdf}} Seminarska 2]
[{{filepath: 03_projekt_VelikaOmrezja_SlavkoZitnik.pdf}} Seminarska 3]

=== Umetna inteligenca ===

Predmet je razdeljen na 3 sklope, vsak traja 5 tednov in vsakega izvaja drug izvajalec. En sklop sestavljajo 2-3x predavanja, 1-2x vaje, zadnji teden pa je namenjen samostojnemu delu, dokončanju in oddaji domačih nalog.

Vsak sklop se zaključi z delno oceno. V vsakem sklopu je nujno doseči oceno najmanj 50%. Končna ocena bo zaokroženo povprečje vseh treh delnih ocen.

[{{filepath:01_UI3_SlavkoZitnik.pdf}} Seminarska naloga 1]

[{{filepath:Dn_planning_SlavkoZitnik.pdf}} Seminarska naloga 2]

[{{filepath:DN_UI_SlavkoZitnik.pdf}} Seminarska naloga 3]

=== Multivariatna analiza ===

Predavanja je izvajala prof. dr. Anuška Ferligoj in zadnje predavanje o strukturnih modelih Germa Coenders.

[{{filepath:00_dn_MA_SlavkoZitnik.docx}} Naloga 0]
[{{filepath:01_dn_MA_SlavkoZitnik.docx}} Naloga 1]
[{{filepath:02_dn_MA_SlavkoZitnik.docx}} Naloga 2]
[{{filepath:03_dn_MA_SlavkoZitnik.docx}} Naloga 3]
[{{filepath:04_dn_MA_SlavkoZitnik.docx}} Naloga 4]
[{{filepath:05_dn_MA_SlavkoZitnik.docx}} Naloga 5]
[{{filepath:SEM_dn_MA_SlavkoZitnik.pdf}} Naloga 6]

=== Raziskovalno delo ===

[{{filepath:R1-program-1_slavko.doc}} Program]
[{{filepath:R2-vmesno_porocilo_I_letnik-1_slavko.pdf}} Vmesno poročilo]
[{{filepath:R3-koncno_porocilo-4_slavko.doc}} Končno poročilo]

== 2nd year (2011/2012) ==

=== Seminar 3 ===

V okviru predmeta boš pripravil kratko, deset minutno ustno predstavitev tvojega tekočega raziskovalnega dela. Predstavitev boš pripravil sam v sodelovanju s tvojim mentorjem. Predstavitev naj obsega:

* Uvod in pregled področja: s katerim raziskovalnim področjem se ukvarjaš? Katera so glavna znanstvena dela na tem področju oz. kakšno je stanje raziskav na tem področju. Sorodna dela, ki podajo stanje na področju, predstavi na kratko, izpostavi seveda del, ki je pomemben za tvoje raziskovanje)?
* Motivacija: zakaj je to področje zanimivo? Kaj raziskuješ oziroma razvijaš ti? Zakaj je to pomembno?
* Metode: katere metode razvijaš? Kaj si že razvil? Kako preveriš, če to, kar si razvil, dobro deluje? Opiši tvoje rezultate.
* Diskusija in zaključek: Si primerjal tvojo rešitev z ostalimi znanimi rešitvami? V čem so prednosti in slabosti tvojo rešitve?

[{{filepath:ZitnikSlavko-clanek.pdf}} Članek]
[{{filepath:ZitnikSlavko.pdf}} Predstavitev]
[http://prezi.com/nqczpedy-zae/seminar-3-iobie/ Prezi link]

=== Seminar 4 ===

[{{filepath:Collie_2nd_submission_SlavkoZitnik.pdf}} Članek]
[{{filepath:Seminar4_prezzi.zip}} Predstavitev]
[http://prezi.com/no2phtphffar/collective-ontology-based-information-extraction-using-probabilistic-graphical-models/ Prezi link]

=== Raziskovalno delo ===

[{{filepath:R1-program_slavko.doc}} Program]
[{{filepath:R2-vmesno-porocilo-ii-letnik-1_slavko.doc}} Vmesno poročilo]
[{{filepath:R3-koncno-porocilo.doc}} Končno poročilo]
[{{filepath:R3-koncno-porocilo-predstavitev-SlavkoZitnik.pdf}} Predstavitev končnega poročila]

=== Tema doktorske disertacije ===

[{{filepath:DR_studij_na_UL_pravila_dopolnjena_izdaja_2012-1.pdf}} Pravilnik UL glede doktorskega študija]
[{{filepath:Napotki-o-pripravi-in-zagovoru-teme.pdf}} Kratka navodila za pripravo teme doktorske disertacije]

[{{filepath:tema-Phd-slavko.pdf}} Prijava teme doktorske disertacije]
[{{filepath:Disposition_slavko.pdf}} Javna predstavitev teme doktorske disertacije]

== 3rd year (2012/2013) ==

Research work going on ...

== 4th year (2013/2014) ==

=== Seminar 5 ===

Zaključna predstavitev doktorskega dela pred končnim zagovorom doktorske disertacije.

Oddan osnutek disertacije pred predstavitvijo: [{{filepath:Thesis-slavko-print_Seminar5.pdf}} Osnutek disertacije]

Predstavitev: [{{filepath:Thesis-slavko-presentation_Seminar5.pdf}} Predstavitev]

=== Zagovor doktorske disertacije ===

Doktorsko delo: [{{filepath:Cover-thesis-slavko.pdf}} Platnica], [{{filepath:Thesis-slavko.pdf}} Disertacija]

Predstavitev na javnem zagovoru: [{{filepath:Thesis-slavko-presentation.pdf}} Predstavitev]

Posnetek javnega zagovora: [https://www.youtube.com/watch?v=3R3Ob5tHL7c Youtube posnetek]

Izvorna koda predlaganih postopkov: [https://bitbucket.org/szitnik/iobie https://bitbucket.org/szitnik/iobie]