Open Source Java Machine Learning Software for Windows

Amine is an Artificial Intelligence Multi-Layer Java Open Source Platform dedicated to the development of various kinds of Intelligent Systems and Agents (Knowledge-Based, Ontology-Based, Conceptual Graph -CG- Based, NLP, Reasoning and Learning, Natural Language Processing, etc.). Ontology, KB can be created and manipulated with various processes. CG theory is used as the main knowledge representation language. Amine provides two languages: PROLOG+CG which extends PROLOG with CG and Amine modules, and SYNERGY which is a visual activation/propagation based language. CGs are considered by SYNERGY as activable/executable graphs. See for more detail: //amine-platform.sourceforge.net/

Highly reusable and extensible Decision-Tree (Max-Gain) framework comprising of comprehensive input-processing and display functionality. Handles nominal, linear, continuous data. For preliminary description, refer - http://sushain.com/blog/archives/

See TTS demo at: http://rslp.racai.ro/index.php?page=tts This is an entirely written in JAVA project which includes a set of tools and methods designed to enable Multilingual Text-to-Speech (TTS) synthesis. We currently support English and Romanian but we will soon train more models and make them available for download. If you want to read more about our other NLP and TTS tools check out http://nlptools.racai.ro.

Drug name recognition and normalisation/grounding to DrugBank ids and standard names. Package provides 2 taggers: 1. DrugTagger - CRF-based with DrugBank presence feature (see feature set for details). 2. DrugnameGazetteer - gazetteer/dictionary-based. Dictionary created from DrugBank.ca database. Both taggers include grounding/normalisation to DrugBank ids and standard names. Feature set: Word, Word-1, Word+1, Word-1_Word, Word_Word+1, DrugBankPresence, POS DrugBankPresence feature indicates the presence of the drug name in the DrugBank. Using CONLL-Evaluation: processed 32065 tokens with 3656 phrases; found: 3251 phrases; correct: 2786. accuracy: 95.25%; precision: 85.70%; recall: 76.20%; FB1: 80.67 Using GATE Corpus Benchmark: Strict: P: 0.65 R: 0.73 F1: 0.69 Lenient: P: 0.74 R: 0.84 F1: 0.78 The details of how to reproduce evaluation, see README. To use standalone version for tagging download DrugExtractionStandalone.tar.gz from Files.

The Edlin toolkit provides a machine learning framework for linear models, designed to be easy to read and understand. The main goal is to provide an architecture and easy to edit working examples of implementations for popular learning algorithms.

FENNIX is a simulator of artificial neural networks written in Java. It allows you to easily describe a complete simulation by using a simple text script language or by adding nodes to a tree of tasks by using the graphical used interface. Moreover, FENNIX is composed of pluggable tools that can be easily modified in order to add new functionalities to the simulator.

Feating constructs a classification ensemble comprising a set of local models. It is effective at reducing the error of both stable and unstable learners, including SVM. For details see the paper at http://dx.doi.org/10.1007/s10994-010-5224-5.

A graphical MatLab framework for estimating the parameters of, modeling and simulating static and dynamic linear and polynomial systems in the errors-in-variables context with the intent of comparing various estimation strategies.

GAME stays for Generic Architecture based on Multiple Experts. Its main purpose is to make easy prototyping, test and release of prediction systems. Released by IASC group, university of Cagliari

The JaCHMM - the Java Conditioned Hidden Markov Model library - is a complete implementation of a CHMM in Java ready to use either on command line or as a module. The JaCHMM is licenced under the BSD licence. It gives an implementation of the Viterbi, Forward-Backward, Baum-Welch and K-Means algorithms, all adapted for the CHMM. JaCHMM is based on the JaHMM and also designed to achieve reasonable performance without making the code unreadable. Consequently, it offers a good way of applying the Conditioned Hidden Markov Model in various tasks, e.g., for scientific or teaching purposes.

LPCforSOS is a machine learning framework with a special focus on structured output spaces and pairwise learning. It supports currently multiclass, ordinal, hierarchical, multi-label and label ranking classification settings.

A Machine Learning and Data Retrieval Framework

This site contains four packages of Mass and mass-based density estimation. 1. The first package is about the basic mass estimation (including one-dimensional mass estimation and Half-Space Tree based multi-dimensional mass estimation). This packages contains the necessary codes to run on MATLAB. 2. The second package includes source and object files of DEMass-DBSCAN to be used with the WEKA system. 3. The third package DEMassBayes includes the source and object files of a Bayesian classifier using DEMass. DEMassBayes.7z has jar file to be used with WEKA and a readme file listing parameters used. The source files are included in DEMassBayes_Source.7z. 4. The four package is MassTER includes source and JAR file to be used with WEKA system..

An implementation of Back Propagation Neuron Network

Neuroph OCR - Handwriting Recognition is developed to recognize hand written letter and characters. It's engine derived's from the Java Neural Network Framework - Neuroph and as such it can be used as a standalone project or a Neuroph plug in.

Feature construction and selection are two key factors in the field of Machine Learning (ML). Usually, these are very time-consuming and complex tasks because the features have to be manually crafted. The features are aggregated, combined or split to create features from raw data. This project makes use of ontologies to automatically generate features for the ML algorithms. The features are generated by combining the concepts and relationships that are already in the knowledge base, expressed in form of ontology.

Sanchay is a collection of tools and APIs for language researchers. It has some implementations of NLP algorithms, some flexible APIs, several user friendly annotation interfaces and Sanchay Query Language for language resources.

Spec is a voice control based on the libraries of Sphinx-4.

Supertagging is a process of statistical lexical disambiguation, preprocessing step to parsing, which assigns LTAG tree categories to the lexical items present in the input sentence. Thus, if the input sentence is in the form of a dependency tree, the task of the supertagger is to assign the most probable TAG family to each node and edge in the dependency tree.

Conversion between different grammar frameworks is of great importance to comparative performance analysis of the parsers developed on them. This tool can convert CCG derivations to PTB trees by using Max Entropy models as well as visualizing the tree graphs. The main technical innovation presented here is the effective conversion method which achieves a F score over 95%.

UnBFuzzy is an API for fuzzy-logic and learning algorithms for fuzzy systems.

Artificial neural network with eigenfaces for face recognition

Moved to https://github.com/quansun/fantail-ml

We moved to GitHub: https://github.com/grobid/grobid