LoginSign Up
← Back to all sparks
B

btm

AI-ASSISTANTS
Velocity0.0

Biterm Topic Models for Short Text

github.com/bnosac/btm ↗

BTM has shipped nothing but compiler and integration compliance since 2020

nlptopic-modelingshort-textr-packagemaintenance
◆Current state
BTM is an R binding to the Biterm Topic Model, aimed at short texts where standard LDA struggles. Its algorithmic surface has not changed in the visible history. Releases since 0.3.3 consist of a fedora-clang self-assignment fix, a terms.data.frame adjustment for compatibility with hardhat's assumptions, clang readability fixes, removal of the C++11 requirement, and documentation NOTEs about itemize.
◆Where it's heading
The package is finished in the sense that matters: the model works and the maintainer keeps it compiling. What movement there is comes from outside — a compiler flag, a CRAN check, another package's expectation about what stats::terms returns. It moves in lockstep with the rest of the bnosac NLP set, which received the same C++11 and packaging cleanups within a day of this one.
◆Prediction
Nothing in the history points at model or interface work, so expect the next release whenever a CRAN check or toolchain change forces one across the sibling packages.

◆Recent moves

  1. 8mo ago

    Clear R CMD check NOTEs about itemize usage

    Documentation markup corrections to silence CRAN check NOTEs. Landed alongside the same sweep across doc2vec and word2vec.

    View source ↗
  2. 3y ago

    clang readability fixes; C++11 requirement dropped

    Applies clang-detected declaration and return-type readability fixes contributed upstream and removes C++11 from SystemRequirements. Compile-time only.

    View source ↗
  3. 5y ago

    Remove unused LazyData; add plot example to README

    Drops a LazyData declaration for a package with no data and adds a plotting example to the README. Housekeeping.

    View source ↗
  4. 5y ago

    terms.data.frame returns existing terms attribute for hardhat

    Changes terms.data.frame to return an existing terms attribute when present rather than always falling through to the default method, working around hardhat's assumption about what stats::terms yields. Narrow, but it is what lets BTM sit inside tidymodels-style pipelines.

    View source ↗
  5. 5y ago

    Fix -Wself-assign on fedora-clang

    A compiler warning fix specific to the fedora-clang check platform.

    View source ↗
  6. 5y ago

    Make example conditional on udpipe availability

    Guards an example against a missing udpipe install — the same fix applied across the bnosac packages that share the dependency.

    View source ↗