| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document Help

Title:Prepoznava štetja kart pri igri blackjack z metodami strojnega učenja : diplomsko delo visokošolskega študijskega programa Informacijska varnost
Authors:ID Berčič, Aljaž (Author)
ID Vrhovec, Simon (Mentor) More about this mentor... New window
Files:.pdf VS_Bercic_Aljaz_2021.pdf (1,52 MB)
MD5: DFD8C6DAE84098FC0A220D04DA4F8203
PID: 20.500.12556/dkum/64ac81cc-4763-4762-a17b-3d0b73b7a273
 
.zip VS_Bercic_Aljaz_2021.zip (175,16 KB)
MD5: 4D97C4CBAF77DE43188426EB0887AACA
PID: 20.500.12556/dkum/adc808f1-93ad-4344-99ca-19927195ab9c
 
Language:Slovenian
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FVV - Faculty of Criminal Justice and Security
Abstract:V tem diplomskem delu smo raziskali igro blackjack, strategije štetja kart, strojno učenje in metode za prepoznavo igralcev, ki štejejo karte. Blackjack je ena izmed najstarejših in najbolj priljubljenih igralniških iger s kartami na svetu. Pravila igre so se skozi čas zelo spreminjala, eden izmed razlogov zato pa je prav gotovo strategija štetja kart in njen razvoj. V diplomskem delu smo tako preverili različno literaturo o igri blackjack in vplivu različnih pravil na samo igro. Raziskali smo različne strategije štetja kart in njihov razvoj. Zaradi hitrega razvoja tehnologije in mobilnih aplikacij je postalo štetje kart dostopno in mnogo lažje za povprečnega igralca. Preverili smo, kako so se igralnice soočile s tem izzivom, saj so vstopali številni igralci, ki so bili opremljenih z znanjem štetja kart. Z uporabo zahtevnejših aplikacij kot je CVCX smo tudi matematično preverili, kako štetje kart, natančneje strategija Hi-Lo, ki je zelo preprosta in popularna, vpliva na igralnice ter koliko lahko igralec, ki šteje karte igralnico oškoduje. Zaradi pomanjkanja raziskav na področju prepoznave igralcev, ki štejejo karte, smo se odločili, da z uporabo metod strojnega učenja – natančneje odločitvenega drevesa, poskušamo identificirati igralce, ki štejejo karte. Zato smo v diplomskem delu natančneje raziskali strojno učenje, algoritme in metode, ki se pri strojnem učenju uporabljajo ter jih uspešno uporabili pri igri blackjack. Odgovorili smo na raziskovalno vprašanje, ali lahko z metodami strojnega učenja prepoznamo igralce, ki štejejo karte. Rezultati so nam pokazali, da jih lahko uspešno prepoznamo. Uspešnost je bila v primeru, ko gledamo samo igralca, ki šteje karte, več kot 80 %. Vendar smo se pri rezultatih soočili z omejitvami, saj smo veliko število osnovnih igralcev napačno klasificirali kot igralca, ki šteje karte. To nas je pripeljalo do zaključkov, da program še ni popoln in je mogočih še veliko nadgradenj saj ne želimo osnovnih igralcev, ki so igralnicam glavni vir prihodka, odsloviti iz igralnice.
Keywords:diplomske naloge, blackjack, štetje kart, strojno učenje, odločitvena drevesa, prepoznava
Place of publishing:Ljubljana
Place of performance:Ljubljana
Publisher:[A. Berčič]
Year of publishing:2021
Year of performance:2021
Number of pages:VII, 44 str.
PID:20.500.12556/DKUM-79535 New window
UDC:004.85(043.2)
COBISS.SI-ID:72578051 New window
Publication date in DKUM:09.08.2021
Views:1399
Downloads:162
Metadata:XML DC-XML DC-RDF
Categories:FVV
:
BERČIČ, Aljaž, 2021, Prepoznava štetja kart pri igri blackjack z metodami strojnega učenja : diplomsko delo visokošolskega študijskega programa Informacijska varnost [online]. Bachelor’s thesis. Ljubljana : A. Berčič. [Accessed 24 March 2025]. Retrieved from: https://dk.um.si/IzpisGradiva.php?lang=eng&id=79535
Copy citation
  
Average score:
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share


Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Link:http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:16.07.2021

Secondary language

Language:English
Title:Blackjack advantage players identification using machine learning methods
Abstract:In this thesis, we made a research on the game of blackjack, card counting strategies, machine learning and its methods for card counting identification. Blackjack is one of the oldest and most popular table card games in the world. The rules for this game were constantly changing over the years. One of the main reasons for this is certainly development of card counting strategies, which has grown in popularity together with the game itself. In thesis, we have checked different literature about blackjack, the influence of different rules and how they affect game itself. We have made a research on different card counting strategies and their development over time. Rapid technology and mobile applications development has made card counting accessible much easier for average player. We have checked how casinos faced those challenges as many players equipped with card counting skill have entered. With using advanced blackjack programs like CVCX, we have mathematically check how card counting, particularly Hi-Lo strategy, which is very simple and popular, affects casinos and how much player, who count cards can harm casinos revenue. Duo to the lack of research in card counting identification field, we decided to use machine-learning methods – decision trees in particular, to try to identify players who count cards. Therefore, in thesis, we did detailed research on machine learning, algorithms and methods, which are used in machine learning and successfully applied them in the game of blackjack. We have answered on research question of whether we can identify card counters using machine-learning methods. The results have showed us that we can successfully identify them. Success rate, when we only look at the players counting cards, was above 80 %. However, we have faced some limitations with results, reason being many basic players been misclassified as card counter. This has lead us to conclusions, that our program is not perfect and could be upgraded, as we don’t want basic players, which are casino’s main source of revenue, to be backed off.
Keywords:blackjack, card counting, machine learning, decision tree, identification


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica