TiMBL, the Tilburg Memory-Based Learner, is a supervised learning toolkit used to classify examples according to their similarity to previously observed cases. It is particularly associated with natural language processing, where the examples often consist of symbolic linguistic features rather than continuous numerical measurements.
Instead of fitting a model that discards its training examples, TiMBL retains an explicit representation of the training data. Classification then compares an unseen example with stored instances, using feature and distance weighting to influence the result. Researchers can examine these decisions and experiment with different learning algorithms, making the software useful for reproducible work with linguistic categories.
TiMBL has been used as a component of established language technology projects, including the Frog Dutch linguistic analysis system. It can also be used for classification tasks outside computational linguistics.
This is free and open source software.
Key Features
- Implements IB1, IB2, IGTree, TRIBL and TRIBL2 learning algorithms.
- Provides k-nearest-neighbour classification with symbolic feature weighting.
- Includes a decision-tree approximation of memory-based classification.
- Supports overlap, modified value difference, cosine and other similarity measures.
- Offers information gain, gain ratio, chi-squared and shared-variance feature weights.
- Includes Levenshtein and Dice per-value similarity calculations.
- Provides leave-one-out evaluation and internal cross-validation.
- Exposes server functionality and an application programming interface.
- Can report nearest neighbours and decision details for analysis.
- Supports multiple CPUs and user-defined example weighting.
- Provides inverse, inverse-linear and exponentially decaying distance weights.
- Allows alternative classification settings to be compared using the same training examples.
- Reports detailed statistics for the learned instance base and classification experiments.
- Cross-platform support – runs on Linux and macOS.
Website: github.com/LanguageMachines/timbl
Support:
Developer: Ko van der Sloot, Antal van den Bosch and contributors
License: GNU General Public License v3.0
TiMBL is written in C++. Learn C++ with our recommended free books and free tutorials.
Related Software
| C++ Natural Language Processing Tools | |
|---|---|
| text2vec | Framework with API for text analysis and natural language processing |
| Moses | Statistical machine translation system |
| TiMBL | Tilburg Memory-Based Learner |
| MITIE | MIT Information Extraction |
| MeTA | Modern C++ data sciences toolkit |
| Colibri Core | Efficient n-gram & skipgram modelling on text corpora |
| CRF++ | Yet Another CRF toolkit |
| BLLIP Parser | Statistical natural language parser |
Read our verdict in the software roundup.
 LinuxLinks helps you discover the best free and open source software for Linux. Explore our extensive collection of recommended applications, including alternatives to popular proprietary software.We also publish software reviews, Linux tutorials, hardware reviews, free programming books and articles covering the wider open source community. Browse our Linux resources to discover more. Know a useful free and open source Linux application we haven't covered? Suggest a program for inclusion. Has your software been featured on LinuxLinks? Display one of our Featured on LinuxLinks badges on your website or GitHub README. |
Please read our Comment Policy before commenting.