Case-Based Reasoning is an algorithm widely used to make an artificial intelligence make decisions in more or less complex situations. The choice is based on the comparison between past events stored in a database which, on the basis of certain characteristics, are recognized as similar or similar. Once the one with the greatest similarity is established, weighing the various characteristics with numerical values, the A.I. will perform an action very similar to the one that solved the problem in that specific situation. Once resolved, the event just occurred and the relative solution are stored in the database, to be used later. The objective of this thesis is to deepen this methodology, starting from how it was born and then describing its structure, observing real application examples through Python codes and finally presenting my initial thesis idea, which led me to choose this subject.
Il Case-Based Reasoning è un'algoritmo molto usato per far prendere decisioni a un'intelligenza artificiale in situazioni più o meno complesse. La scelta si basa sul confronto tra eventi passati memorizzati all'interno di un database, che, in base a determinate caratteristiche sono riconosciuti come simili o affini. Una volta che viene stabilito quello con maggiore somiglianza, pesando le varie caratteristiche con valori numerici, l'A.I. eseguirà un azione molto simile a quella che aveva risolto il problema in quella specifica situazione. Una volta risolto, l'evento appena avvenuto e la relativa soluzione vengono memorizzati all'interno del database, per poter essere utilizzata in seguito. L'obiettivo di questa tesi è quello di approfondire questa metodologia, partendo da come è nata per poi descrivere la sua struttura, osservando esempi di applicazione reali attraverso codici Python e infine presentando la mia idea iniziale di tesi, che mi ha portato a scegliere questo argomento.
Case Based Reasoning: teoria ed applicazioni
VIOLA, SAMUELE
2022/2023
Abstract
Case-Based Reasoning is an algorithm widely used to make an artificial intelligence make decisions in more or less complex situations. The choice is based on the comparison between past events stored in a database which, on the basis of certain characteristics, are recognized as similar or similar. Once the one with the greatest similarity is established, weighing the various characteristics with numerical values, the A.I. will perform an action very similar to the one that solved the problem in that specific situation. Once resolved, the event just occurred and the relative solution are stored in the database, to be used later. The objective of this thesis is to deepen this methodology, starting from how it was born and then describing its structure, observing real application examples through Python codes and finally presenting my initial thesis idea, which led me to choose this subject.File | Dimensione | Formato | |
---|---|---|---|
Viola_Samuele.pdf
accesso aperto
Dimensione
705.02 kB
Formato
Adobe PDF
|
705.02 kB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/52615