titan>> Titanic-Investigation-and-Machine-Learning-from-Disaster>> 返回
项目作者： ankitrhode

项目描述 ：
Investigation of passenger's features against survival on Titanic and Machine Learning on Titanic dataset
高级语言： Jupyter Notebook
项目地址: git://github.com/ankitrhode/Titanic-Investigation-and-Machine-Learning-from-Disaster.git


Titanic-Investigation-and-Machine-Learning-from-Disaster

TITANIC /https://public-media.smithsonianmag.com/filer/ae/fc/aefc1fc2-e241-4a17-b73b-9ad6518427cb/stower_titanic.jpg)

  • Goal of this Notebook:

    • Investigation - whether passenger’s features correlate with survival?
    • Analysis of Titanic dataset.
    • Finding the best working classifier for Titanic dataset.

  • This Notebook will show basic examples of:

    • Investigation - whether passenger’s features correlate with survival?

      • Importing libraries
      • Importing data with Pandas
      • Feature engineering
      • Determine passenger’s features to test against survival¶

    • Analysis of Titanic dataset

      • Plot/graph variables and mathematically exploration

    • Machine Learing on Titanic dataset

      • Importing machine learning libraries
      • Data cleaning
      • Feature extraction
      • Classification algorithms
      • Visualization

  • Dependencies

    • Numpy: It provides a high-performance multidimensional array and basic tools to compute with and manipulate these arrays.
    • Pandas: For data manipulation and analysis.
    • SciKit-Learn: Python machine learning library which features various classification, regression and clustering algorithms.
    • Matplotlib: Python Plotting.
    • Seaborn: Python’s Statistical Data Visualization Library.
    • Scipy: It provides a large number of functions that operate on numpy arrays and are useful for different types of scientific and engineering applications.

  • Overview of Titanic dataset

    • survival: Survival; 0 = No, 1 = Yes
    • pclass: Ticket class, a proxy for socio-economic status (SES); 1 = 1st (Upper), 2 = 2nd (Middle), 3 = 3rd (Lower)
    • sex: Sex of passenger
    • Age: Age in years
    • sibsp: # of siblings / spouses aboard the Titanic
    • parch: # of parents / children aboard the Titanic
    • ticket: Ticket number
    • fare: Passenger fare
    • cabin: Cabin number
    • embarked: Port of Embarkation; C = Cherbourg, Q = Queenstown, S = Southampton

  • Conclusions

    • Investigation: Which passenger characteristics correlate with survival?

      • Passenger gender correlates with survival: Females were more likely to survive than males.
      • Passenger class correlates with survival: The higher the passenger’s class, the more likely that he or she survived.

    • Machine Learning on Titanic dataset

      • GradientBoostingClassifier can be used to find insights on the Titanic dataset.