Final Exam: Predictive Analytics

WHAT YOU WILL LEARN

• 

  Read in a network attack detection dataset to pandas
  

  create a column for the cyberattack type
  

  visualize network cyberattack data and view the unique values
  

  predict cyberattacks using heatmap and pie charts
  

  perform one-hot encoding on cyberattack data and view the results
  

  view the performance of an overfit cyberattack prediction model
  

  use a chi-square test to perform feature selection
  

  select features for the model using the chi-square test
  

  view the dataset profile and create a pipeline
  

  analyze a machine failure prediction dataset
  

  perform preparations on machine failure prediction data
  

  view the performance of a machine failure prediction model
  

  use smote to improve the performance of a machine failure prediction model
  

  standardize data using a component
  

  create and configure a decision forest model for machine failure prediction
  

  compare the performance of the logistic regression and decision forest models
  

  perform hyperparameter tuning on a machine failure prediction model and view the results
  

  perform preprocessing on data
  

  visualize data and remove outliers
  

  use ridge regression, knn, decision trees, extra tree regressors, and random forests to predict walmart sales
  

  perform cross-validation and feature selection on a clv prediction model
  

  perform feature selection on clv prediction model
  

  perform feature selection using recursive feature elimination (rfe), lasso regression, and support vector regression (svr) on a clv prediction model
  

  view correlations and relationships in e-commerce data
  

  split and encode e-commerce data for machine learning
  

  import and process marketing campaign data
  

  create a pipeline for performing classification
  

  view the performance of a marketing response prediction model
  

  calculate recency, frequency, and monetary (rfm) value of customers
  

  standardize and normalize rfm data for clustering
• 

  perform k-means clustering on rfm data
  

  perform agglomerative clustering on rfm data
  

  perform dbscan clustering on rfm data
  

  encode data for performing market basket analysis
  

  perform market basket analysis
  

  create an azure machine learning workspace
  

  read in data to pandas data frames and explore the data
  

  visualize data attributes
  

  view attribute relationships in the data
  

  select the most important attributes for classification
  

  classify data by type using machine learning models
  

  visualize data about a crop's climatic conditions
  

  transform crop data for clustering
  

  optimize the clustering of climatic data
  

  set up a climatic condition clustering model
  

  fine-tune the clustering of climatic data
  

  perform transformations on blueberry yield data
  

  predict blueberry yield using linear regression
  

  identify attributes that help with yield prediction
  

  apply data cleaning to numeric fields in a dataset
  

  handle missing values in categorical fields in a dataset
  

  group numeric attributes into bins so that they can be treated as categorical fields
  

  register cleaned and processed data as a dataset
  

  split a dataset into train and test sets
  

  create and use a real-time inferencing pipeline
  

  deploy a kidney disease model to an azure container and consume it
  

  build a pipeline using a template densenet model to detect tumor types
  

  configure and run an image classification pipeline to detect tumors
  

  configure the model's parameters and evaluate the improved performance on the test data
  

  perform image classification using a resnet model