鼠鼠学PYTHON

《学习笔记模式》

Here are some guidelines explaining the scope that the final exam might cover.

1. Data Processing (15%):
   • File Handling with Python
     • Read files (e.g., txt, CSV);
     • Write to files (e.g., txt, CSV);
   • Scientific Computing Library: using numpy module.
     • Array Creation, Shaping, and Transposition
       • Array Creation Routines
       • Shaping
     • Transposition
     • Indexing & Slicing
     • Mathematical Operation
     • Statistics & Linear Algebra   
   • Python Data Analysis Library: using pandas module.
     • Create DataFrame & Series;
     • Indexing & Slicing;
     • Calculation, Joining and Concatenation;
     • Some advanced pandas operations
2. Data Visualization (5%): line plot, scatter plot, bar plot, histogram.
   • Visualization using Matplotlib;
   • Visualization using Seaborn;
3. Web Scraping (10)
   • Navigate a tree of an HTML doc using BeatifulSoup;
4. Time Series Analysis (10%): using statsmodels.tsa.
   • Create time series;
   • Check whether the time series is stationary;
   • Exponential smoothing;
   • Component Decomposition;
   • Autocorrelation function (ACF) & Partial autocorrelation function (PACF)
   • Model Diagnostics & Model Selection of ARIMA
5. Machine Learning (30%): using scikit-learn library.
   • Supervised Learning
     • Classification: Basic concepts of Logistic Regression (sigmoid function), Support Vector Machines (kernel), K-Nearest Neighbor (weight), Naive Bayes, Decision Tree, Random Forests.
       • Binary Classification
       • Multiclass Classification
         • One-versus-All (OvA) Strategy
         • One-versus-One (OvO) Strategy
     • Regression: OLS, Ridge Regression, LASSO;
       • Linear Regression
       • Polynomial Regression
   • Unsupervised Learning: Basic concepts of K-Means and PCA
     • Clustering
       • K-Means
         • Finding the Optimal Number of Clusters;
         • Inertia;
         • Silhouette Score    
     • Decomposition
       • Principle Component Decomposition
         • Variance Explained
   • Training ML Models;
     • Prepare data for model training;
     • Training, Validating, and Testing (e.g., Cross-validation);
     • Fine-tuning the model (e.g., GridSearchCV);
   • Evaluate ML Models;
     • Performance measures:
       • Classification: Accuracy, Confusion Matrix, Precision, Recall, F1 Score, ROC
       • Regression: MSE, RMSE, MAE, RMAE
   • Make Predictions;
   • Deal with Missing Values;
   • Feature Scaling;
   • Handling Text and Categorical Attributes: encoder;
6. Natural Language Processing (20%)
   • Simple Regular Expression using re module;
   • Tokenization and Stemming using NLTK;
   • Topic analysis using Gensim:
     • Latent Dirichlet Allocation (LDA);
     • Basic concepts: Document, Corpus, Dictionary, etc.
     • Bag-of-Word model;
   • Analyze text using SpaCy:
     • Tokenization;
     • Part of Speech (POS) Tagging
     • Named Entity Recognition (NER)
   • Others:
     • Feature representation/Vectorization: Count vectorizer and TF-IDF vectorizer using scikit-learn;
     • Text classification: such as using Naive Bayes to classify text;
     • Sentiment Analysis using nltk.sentiment.vader;
7. Network Analysis (10%): using NetworkX module.
   • Basic concepts of network analysis: node, edge, graph, etc.
   • Creating a graph;
   • Network Properties
   • Graph visualization
   • Node2vec embedding;

可用以下模式搜集Python练习题，并附上相关知识点。该系列将不断更新。

Question 1:

----
##

Question 2:  

----
##

Question 3:   

----
##