ECTS - University of Pisa
Gli studenti che completeranno con successo l'insegnamento avranno una solida conoscenza delle principali tecniche per pre-processare i dati, frequent pattern mining, frequent sequential pattern mining, graph mining, classificazione, predizione, clustering, outlier detection. Questa conoscenza permetterà loro di affrontare ogni tipo di problema inerente il data mining e di identificare la tecnica più adatta per risolverlo.
The students who successfully complete the course will have a solid knowledge of the main techniques used in data preprocessing, frequent pattern mining, frequent sequential pattern mining, graph mining, classification, prediction, clustering and outlier detection. This knowledge will allow them to tackle each type of data mining problem and to identify the most suitable technique for solving it.
Durante la verifica delle conoscenze, gli studenti devono dimostrare di aver appreso le diverse tecniche insegnate durante lo svolgimento del corso e devono essere capaci di identificare la soluzione più adatta per problemi di data mining specifici.
I metodi sono:
Ulteriori informazioni: allo studente è richiesto di sviluppare un progetto in cui vengono utilizzate tecniche di data mining. I risultati del progetto vengono discussi durante una presentazione.
During the oral exam the student must be able to demonstrate his/her knowledge of the course material and be able to identify the most suitable solutions for specific data mining problems.
Methods:
Further information:
The student is requested to develop an application employing some data mining technique. The results of the project are described in a report and discussed during the presentation of the project.
Al termine del corso,
At the end of the course,
Lo studente dovrà preparare e presentare una relazione scritta che riporti i risultati dell'attività di progetto
The student will have to prepare and present a report, which describes the development of the project and the obtained results
Lo studente potrà acquisire un metodo per affrontare problemi di data mining e per selezionare le migliori soluzioni da adottare
The student will acquire a method to deal with data mining problems and to select the most effective solution to be adopted
Durante le sessioni di laboratorio saranno valutati il grado di accuratezza e precisione delle attività svolte dallo studente
Durante lo sviluppo del progetto saranno verificate le modalità di gestione e organizzazione delle fasi progettuali
During the laboratory sessions the degree of accuracy and precision of the activities carried out by the student will be evaluated
During the development of the project, the procedures for managing and organizing the project phases will be verified
Conoscenze di base di matematica
Conoscenze di linguaggi di programmazione
Basic knowledge of mathematics
Knowledge of programming languages
Le lezioni verranno svolte frontalmente con l'ausilio di slide
Le esercitazioni verranno svolte in laboratorio con l'ausilio di slide e esempi di programmazione.
Durante il corso, verrà sviluppato dallo studente un progetto che costituirà parte della valutazione finale
L'intero corso è tenuto in Inglese
Delivery: face to face
Learning activities:
Attendance: Advised
Teaching methods:
Data Preprocessing: data cleaning, integration, reduction, transformation and discretization.
Frequent pattern mining: basic concepts, A-priori algorithm, Pattern-Growth approach, vertical data format, pattern evaluation methods, constraint-based frequent pattern mining, colossal pattern.
Classification: basic concepts, decision tree induction, Bayes classification methods, rule-based classification, lazy learners, techniques for improving accuracy, model evaluation and selection.
Clustering: basic concepts, partitioning methods, hierarchical methods, density-based methods, grid-based methods, model evaluation and selection, clustering with constraints.
Outlier detection: statistical, proximity-based, clustering-based and classification-based approaches.
Sequential Pattern Mining: basic concepts, AprioriAll, AprioriSome, AprioriDynamicSome
Graph Mining: basic concept, geodesic distance, SimRank, Density-based approaches to graph clustering.
Time series: basic concepts and machine learning for forecasting.
Text mining: basic concepts, text classification and case studies.
Trustworthy AI: basic concepts and current solutions.
Distributed frameworks: basic concepts, Hadoop, MapReduce paradigm, Spark, some examples of data mining algorithms implemented by using MapReduce
Python programming language: Introduction, Jupyter Notebook environment and Machine Learning libraries
Data Preprocessing: data cleaning, integration, reduction, transformation and discretization.
Frequent pattern mining: basic concepts, A-priori algorithm, Pattern-Growth approach, vertical data format, pattern evaluation methods, constraint-based frequent pattern mining, colossal pattern.
Classification: basic concepts, decision tree induction, Bayes classification methods, rule-based classification, lazy learners, techniques for improving accuracy, model evaluation and selection.
Clustering: basic concepts, partitioning methods, hierarchical methods, density-based methods, grid-based methods, model evaluation and selection, clustering with constraints.
Outlier detection: statistical, proximity-based, clustering-based and classification-based approaches.
Sequential Pattern Mining: basic concepts, AprioriAll, AprioriSome, AprioriDynamicSome
Graph Mining: basic concept, geodesic distance, SimRank, Density-based approaches to graph clustering.
Time series: basic concepts and machine learning for forecasting.
Text mining: basic concepts, text classification and case studies.
Trustworthy AI: basic concepts and current solutions.
Distributed frameworks: basic concepts, Hadoop, MapReduce paradigm, Spark, some examples of data mining algorithms implemented by using MapReduce
Python programming language: Introduction, Jupyter Notebook environment and Machine Learning libraries
Slides del corso
Libro: J. Han and M. Kamber. Data Mining: Concepts and Techniques. Morgan Kaufmann, 4th ed., 2022
Articoli sui differentio algoritmi descritti durante le lezioni forniti dal docente
Slides of the lectures
Recommended book: J. Han and M. Kamber. Data Mining: Concepts and Techniques. Morgan Kaufmann, 4th ed., 2022.
Papers on the different algorithms described during the course provided by the teacher
L'esame è composto dalla discussione del progetto e una prova orale.
La discussione del progetto viene tipicamente tenuta qualche giorno prima dell'esame orale. Il candidato deve presentare come il progetto è stato sviluppato, motivare le sue scelte progettuali e discutere i risultati ottenuti. Il progetto viene valutato positivamente se il candidato mostra di aver seguito un approccio corretto e di aver valutato in modo critico le possibili soluzioni, scegliendo la più appropriata
La prova orale consiste in un colloquio tra il candidato e il docente su alcune domande che potrebbero anche assegnate in forma scritta al candidato.
La prova orale è superata se il candidato mostra padronanza degli argomenti trattati, si esprime in modo chiaro e con terminologia corretta, mostra capacità di analisi e sintesi.
The assessment method consists of a presentation of the project and an oral exam.
The presentation of the project will be hold some days before the oral exam. The candidate has to present how the project has been developed, to justify the design choices and critically discuss the obtained results. The project is positively evaluated if the candidate shows to have followed a correct approach and to have critically evaluated the possible solutions, choosing the most appropriate
The oral exam consists of a conversation between the candidate and the teacher on some questions that are assigned in writing to the candidate.
The oral exam is over if the candidate shows mastery of the topics covered, is expressed clearly and with correct terminology, shows the ability of analysis and synthesis.
