Aperçu des sections

  • Généralités

    Tout replier Tout déplier

  • Contact sheet

    Enseignant de la matière :Sihem Benfriha

    Contact : benfriha.sihem@univ-tlemcen.dz

     Disponibilité(horaire et lieu): Mardi 8h:30-10h département informatique 

    Coefficient :3

    Crédits :5

     Volume horaire global : 63h

     Volume horaire de travail personnel requis/semaine: 9h/semaine

  • Objectives

     

    • Recall the essential Python libraries and tools (pandas, numpy, matplotlib, seaborn, scikit-learn) used throughout the data mining workflow.
    • Explain the complete data mining process from data collection, through preprocessing, to model building and deployment .
    • Use Python in Jupyter Notebook to collect, explore, clean, transform, and model real-world datasets, applying appropriate functions and algorithms at each stage.
    • Compare data characteristics, preprocessing techniques, and model outputs to identify the most suitable approach for a given problem.
    • Critically assess the quality of data, the effectiveness of preprocessing choices, and the performance of trained models, justifying decisions with appropriate metrics.
    • Implement a complete, end-to-end data mining pipeline in Python  from raw data to a deployed model/application  capable of solving a real-world problem.


  • Pre-requisites

    - Basic computer literacy (using a computer, keyboard, file management)

    - Fundamental mathematical skills (basic arithmetic operations: addition, subtraction, multiplication, division) 

    - Basic logical thinking (understanding sequences of steps, simple problem-solving

  • Global Plan

    Chapter 1 : Data Collection and Exploration . 

    Chapter 2: Data Preprocessing .

    Chapter 3 : Model Building and Deployment .

  • Meeting with your teacher

  • Chapter 1 : Data Collection and Exploration

      Objectives

    •  List the core Python libraries used for data mining (pandas, numpy, matplotlib, seaborn) and recall basic Jupyter Notebook commands.
    •  Explain the purpose of functions such as .head(), .info(), .describe(), and .shape in understanding a dataset's structure.
    •  Load datasets from various sources (CSV, Excel, API, SQL database) into a pandas DataFrame using Python.
    •  Examine a dataset's distributions, correlations, and outliers using visualizations (histograms, boxplots, scatter plots, heatmaps).
    •  Assess the quality and relevance of a collected dataset by identifying missing values, inconsistencies, and anomalies.

  • Chapter 2: Data Preprocessing

    Objectives

    • Recall the Python functions used for handling missing values and encoding (isnull(), fillna(), dropna(), LabelEncoder, OneHotEncoder).
    • Explain the difference between normalization and standardization, and when each should be applied.
    • Implement preprocessing techniques in Python — handling missing data, encoding categorical variables, scaling numerical features, and splitting data (train_test_split).
    •  Compare the effects of different preprocessing techniques (e.g., MinMaxScaler vs. StandardScaler) on the dataset's distribution.
    • Justify the choice of a specific preprocessing pipeline based on the dataset's characteristics and the intended model.
    •  Build a complete, reusable data preprocessing pipeline in Python using scikit-learn's Pipeline module.

  • Exit Test

    • Icône Test
      Learning Validation Test
      Ouvert : jeudi 20 août 2026, 11:19
      Se termine : dimanche 20 septembre 2026, 11:19

      The test score will be included in the overall average.

  • References

    1.Han, J., Pei, J., & Kamber, M. (2011). Data mining: Concepts and techniques (3rd ed.). Morgan Kaufmann Publishers
    2.Linoff GS, Berry MJ. Data mining techniques: for marketing, sales, and customer relationship management. John Wiley & Sons; 2011 Mar 23.
    4.Han, J., & Kamber, M. (2006). Data mining: Concepts and techniques (2nd ed.). Morgan Kaufmann Publishers. 
    5.Grossman, R. L., & Horn, W. (2004). A survey of data mining software tools. SIGKDD Explorations, 6(1), 20-33. 
    6.Tan, P. N., Steinbach, M., & Kumar, V. (2005). Introduction to data mining (1st ed.). Addison    Wesley. 
    7. Aggarwal, C. C. (2015). Data mining: The textbook (1st ed.). Springer.