Budapest University of Technology and Economics, Faculty of Electrical Engineering and Informatics

    Belépés
    címtáras azonosítással
    vissza a tantárgylistához   nyomtatható verzió    

    Advanced Data Analysis Methods Laboratory

    A tantárgy neve magyarul / Name of the subject in Hungarian: Haladó adatelemzési módszerek labor

    Last updated: 2024. február 19.

    Budapest University of Technology and Economics
    Faculty of Electrical Engineering and Informatics
    Course ID Semester Assessment Credit Tantárgyfélév
    VITMMB10   0/0/3/f 5  
    3. Course coordinator and department Dr. Toka László,
    4. Instructors Bálint Pál Gyires-Tóth, associate professor, TMIT
    Péter Orosz, associate professor, TMIT
    Bence Bolgár, assistant professor, MIT
    Dániel Hadházi, PhD student, MIT
    5. Required knowledge Data Science, Artificial Intelligence, Data Analysis, Statistics, Probability Theory
    6. Pre-requisites
    Kötelező:
    TárgyTeljesítve("BMEVITMMA19") VAGY
    TárgyTeljesítve("BMEVITMMA06")

    A fenti forma a Neptun sajátja, ezen technikai okokból nem változtattunk.

    A kötelező előtanulmányi rend az adott szak honlapján és képzési programjában található.

    Ajánlott:
    Mandatory: Deep Learning
    Recommended: Databases, Artificial Intelligence
    7. Objectives, learning outcomes and obtained knowledge The aim of the course is to deepen theoretical knowledge and practical skills acquired in the Data Science and Artificial Intelligence specialization through the execution of a specific data mining project.
    8. Synopsis
    1. Selection and interpretation of the data mining task, project planning, and documentation of the evaluation criteria for future solutions. Subsequently, complete data mining cycles and redefine them by evaluating the following work stages:

    2. Data Preparation (selection of the database and data format, data cleansing, etc.)
    3. Data Visualization and Analysis (correlation analysis, explanatory variable selection, data transformations, etc.)
    4. Generation of Machine Learning Models (model selection, hybrid, deep learning, etc.)
    5. Evaluation of Machine Learning Models (metric selection, bootstrapping, improving results, hyperparameter tuning, applying boosting, etc.)
    6. Practical application of the generated data mining process (deployment to the cloud, ethical considerations, data protection).
    9. Method of instruction The independent solution of programming tasks related to the milestones defined in the curriculum will be assigned as homework, with the presentation of solutions during the laboratory sessions.
    10. Assessment
    The semester work is organized along the milestones specified in this thematic plan every two weeks. The requirement is to successfully complete at least 4 out of 6 milestones within the specified time, meaning that delaying 2 milestones during the semester is permitted. The final grade for the semester is calculated based on the results of the 6 milestones and the end-of-year report grade. The weekly schedule is as follows:

    1. Introduction, presentation of available tasks, task assignment (attendance required)
    2. Online consultation opportunity during scheduled hours
    3. M1: Presentation of project plan, data preparation plan (attendance required)
    4. Online consultation opportunity during scheduled hours
    5. M2: Presentation of data preparation, data visualization plan (attendance required)
    6. Online consultation opportunity during scheduled hours
    7. M3: Presentation of data visualization, ML models plan (attendance required)
    8. Online consultation opportunity during scheduled hours
    9. M4: Presentation of ML models, ML model evaluation plan (attendance required)
    10. Online consultation opportunity during scheduled hours
    11. M5: Presentation of ML model evaluation, deployment plan (attendance required)
    12. Online consultation opportunity during scheduled hours
    13. M6: Presentation of application (attendance required)
    14. Opportunity for make-up sessions
    11. Recaps In case of missing a maximum of 2 milestones, it is mandatory to complete the outstanding results by the deadline of the next milestone (or, in the case of the 6th milestone, by the penultimate week of the semester).
    12. Consultations Depending on the availability of the owner of the chosen data mining task (see Course Instructors), preferably once a week during a fixed time slot, online communication will be arranged.
    13. References, textbooks and resources

    [1] Jiawei Han, Micheline Kamber, Jian Pei: Data Mining Concepts and Techniques (Third Edition), 2012, https://myweb.sabanciuniv.edu/rdehkharghani/files/2016/02/The-Morgan-Kaufmann-Series-in-Data-Management-Systems-Jiawei-Han-Micheline-Kamber-Jian-Pei-Data-Mining.-Concepts-and-Techniques-3rd-Edition-Morgan-Kaufmann-2011.pdf

    [2] scikit-learn, 2024, https://scikit-learn.org/

    [3] pandas, 2024, https://pandas.pydata.org/
     
    [4] TensorFlow, 2024, https://www.tensorflow.org/
    14. Required learning hours and assignment
    Contact hours42
    Preparing for classes18
    Preparing for test
    Preparing the homework90
    Learning textbook
    Preparing for exam
    Total150