Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

Potential roadmap items 2017+:Some potential roadmap ideas for 2018+ below.  For more details please see the JIRAs

Predictive models

  • Multi-class SVM 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1037
  • Mixed effects modeling 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-987
  • Gradient boosted machines
  • Novelty detection using 1-class SVM  MADLIB-990
  • Mixed effects modeling  MADLIB-987
  • Factorization machines
  • k-nearest neighbors (kNN)  MADLIB-927
  • Geographically Weighted Regression (GWR) MADLIB-1023
  • MCMC Probit and Logit regression
  • Gaussian Mixture Model using Expectation Maximization (EM) algorithm algorithm 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-410
  • Multi-layer Perceptron
  • Hierarchical clustering
  • Neural nets and deep learning

Graph

  • Single source shortest path  MADLIB-992
  • All pairs shortest path
  • One mode projection (converting a bi-partitite graph of user-item graph to user-user or item-item graph)
  • Connected components
  • Page rank MADLIB-1069
  • Graph cut
  • Centrality measures like betweenness and closeness
  • Triangle counting
  • Minimum spanning tree
  • Graph diameter

Utilities

Usability

  • Hierarchical clustering
  • k-NN improvements 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1061
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1181
  • Deep learning

Graph

  • Personalized PageRank 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1084
  • Betweenness centrality 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1121
  • Graph cut 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1074
  • Triangle counting 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1125
  • Minimum spanning tree
  • Eigenvector centrality 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1123
  • APSP performance improvements 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1155

Utilities

  • Balanced datasets 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1168

    Summary - add more statistics 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1167
  • Anonymization  
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-911
  • URI tools  
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-910

Usability

  • Expand coverage for PivotalR
  • Expand coverage for PivotalR
  • Expand coverage for PMML export??? (or perhaps switch to PFA???)
  • Interface improvement and consistency for 2.0 release (does not need to be backward compatible)Implement an interface using named parameters 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1157
  • Python API

Performance and scalability

  • Mini-batching 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1048
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1037
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1200
  • Work around PostgreSQL 1 GB field size limit  MADLIB-991
  • GPU support
  • Improve decision tree and random forest performance for run-time and memory usage MADLIB-1057 MADLIB-976

Platform

  • usage 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1057
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-976

Platforms and Frameworks

  • PostgreSQL 10 support 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1185
  • Support modern versions of gcc 
    Jira
    serverASF JIRA
    serverId5aa69414-a9e9-3523-82ec-879b028fb15b
    keyMADLIB-1025
    Support for PostgreSQL 9.5 and 9.6  MADLIB-944
  • Tensorflow support, or another deep learning framework
  • GPU support