项目实训营
header-img

获得数据科学项目经验

Tigerair 数据科学项目实训营

数据科学专家指导项目,8 周获得数据科学项目经验

header-img
feature增加项目经验
feature数据科学实践
feature导师每周辅导
feature全面的技术栈
feature增加项目经验
feature数据科学实践
feature导师每周辅导
feature全面的技术栈

课程大纲

    入营欢迎会
    Welcome

    Enhancing Airline Passenger Experience

    Objective: The primary goal is to leverage the "Airline Passenger Satisfaction" dataset to uncover insights into what factors contribute most to passenger satisfaction and dissatisfaction. The project aims to predict passenger satisfaction levels based on various service aspects provided by the airline.

    Context: In a competitive airline industry, understanding and improving passenger satisfaction is crucial for retaining customers and enhancing service quality. Airlines strive to identify key factors that influence passenger experience and satisfaction. This project will enable an airline to strategically invest in areas that significantly impact passenger satisfaction, thereby improving overall service quality and competitive advantage.

    Challenge: Students will analyze the dataset to identify patterns and correlations between different service aspects (such as inflight wifi service, seat comfort, and cleanliness) and overall passenger satisfaction. They will develop a predictive model to forecast a passenger's satisfaction level based on these features. The model's accuracy and insights will guide the airline in prioritizing service improvements and personalizing the passenger experience.

    Deliverables:

    • An exploratory data analysis (EDA) report highlighting key factors affecting passenger satisfaction.
    • A predictive model with an evaluation of its performance.
    • Recommendations for the airline on improving passenger satisfaction based on the analysis.

    This project will not only help students apply their data science skills in a real-world context but also contribute to enhancing the airline's service quality by understanding and addressing passenger needs and preferences.

    项目准备
    Data science problem identification and data investigation

    Introduction to Basic Machine Learning Concepts:

    • Overview of machine learning (ML) and its impact on solving real-world problems.
    • Distinction between supervised, unsupervised, and reinforcement learning.
    • Key terminology: features, models, training, and validation.

    Understanding the Business Problem:

    • Techniques for effective communication with stakeholders to understand business objectives.
    • Identifying key performance indicators (KPIs) that align with business goals.

    Transforming Business Problems into Data Science Problems:

    • Strategies for breaking down complex business challenges into manageable data science tasks.
    • Examples of translating common business objectives into specific analytical questions.

    Data Investigation:

    • Steps for initial data exploration, including data quality assessment and preliminary analysis.
    • Importance of understanding the data's context, structure, and potential biases.
    • Techniques for visual data exploration to uncover patterns, trends, and anomalies.
    数据处理
    Exploratory data analysis
    • Variation of Variables: Understand how variables differ among themselves, including range, central tendency, and dispersion measures.
    • Missing Data: Identify and handle missing values through imputation, deletion, or estimation techniques.
    • Covariation: Explore relationships between variables using correlation coefficients, scatter plots, and cross-tabulations.
    • Visualization: Employ visual tools like histograms, box plots, scatter plots, and heat maps to uncover patterns, trends, and outliers in the data.
    Data preprocessing
    • Handling Missing Values: Techniques to detect and treat missing data, such as imputation or removal.
    • Removing Duplicates: Identifying and eliminating duplicate records to ensure data quality.
    • Understanding Data Types: Recognizing and converting data types for proper analysis, including categorical, numerical, and text data.
    • Addressing Data Inconsistency: Standardizing values to resolve inconsistencies in data, ensuring uniformity across datasets.
    特征选择
    Feature engineering
    • Feature Selection: Techniques to identify and select the most relevant features for your model.
    • Creating New Features: Strategies for generating new features from existing data to enhance model performance.
    • Dimension Reduction: Methods like Principal Component Analysis (PCA) to reduce the number of variables, simplifying the model without losing significant information.
    模型选择
    ML model building
    • Advanced ML Concepts: Explore more sophisticated machine learning concepts beyond the basics.
    • Model Selection: Learn how to choose the appropriate model based on the problem type and data characteristics.
    • Implementation: Practical steps for implementing selected models, including training, tuning, and validation processes.
    超参数优化
    Hyperparameter tuning
    • Grid Search: A method to systematically work through multiple combinations of parameter tunes, cross-validating as it goes to determine which tune gives the best performance.
    • Random Search: Involves randomly selecting combinations of parameters to find the best solution for the built model more quickly than the exhaustive grid search method.
    • Bayesian Optimization: A more efficient approach that uses probability to find the minimum or maximum of a function. It builds a probabilistic model of the function and uses it to select the most promising parameters to evaluate in the true objective function.
    模型评估
    Model evaluation
    • Split/Cross-Validation (CV): Techniques to partition data into subsets; training the model on one subset and validating it on another to ensure it generalizes well to new data.
    • Metrics: Different metrics for evaluating model performance, such as accuracy, precision, recall, F1 score for classification tasks, and MSE, RMSE, MAE for regression.
    • Bias/Variance Trade-off: Understanding the balance between bias (error from erroneous assumptions) and variance (error from sensitivity to small fluctuations in the training set) to improve model generalization.
    数据 Pipeline
    ML pipeline
    • Preprocessing Pipeline: Steps to clean and prepare your data for modeling.
    • Feature Engineering Pipeline: Techniques to select, modify, or create new features.
    • Training Pipeline: The process of training your model with the prepared data.
    • Scoring Pipeline: How to apply the model to new data to make predictions.
logo

Follow Us

linkedinfacebooktwitterinstagramweiboyoutubebilibilitiktokxigua

We Accept

/image/layout/pay-paypal.png/image/layout/pay-visa.png/image/layout/pay-master-card.png/image/layout/pay-stripe.png/image/layout/pay-alipay.png

地址

Level 10b, 144 Edward Street, Brisbane CBD(Headquarter)
Level 8, 11 York st, Wynyard, Sydney CBD
Business Hub, 155 Waymouth St, Adelaide SA 5000

Disclaimer

footer-disclaimerfooter-disclaimer

JR Academy acknowledges Traditional Owners of Country throughout Australia and recognises the continuing connection to lands, waters and communities. We pay our respect to Aboriginal and Torres Strait Islander cultures; and to Elders past and present. Aboriginal and Torres Strait Islander peoples should be aware that this website may contain images or names of people who have since passed away.

匠人学院网站上的所有内容,包括课程材料、徽标和匠人学院网站上提供的信息,均受澳大利亚政府知识产权法的保护。严禁未经授权使用、销售、分发、复制或修改。违规行为可能会导致法律诉讼。通过访问我们的网站,您同意尊重我们的知识产权。 JR Academy Pty Ltd 保留所有权利,包括专利、商标和版权。任何侵权行为都将受到法律追究。查看用户协议

© 2017-2024 JR Academy Pty Ltd. All rights reserved.

ABN 26621887572