The Graphical Models Toolkit (GMTK) is a toolkit for rapidly prototyping statistical models using dynamic graphical models (DGMs) and dynamic Bayesian networks (DBNs). It can be used for speech and language processing, bioinformatics, activity recognition, and any time series application. It features exact and approximate inference, many built-in factors including dense, sparse, and deterministic conditional probability tables, native support for ARPA backoff-based factors and factored language models, parameter sharing, gamma and beta distributions, dense and sparse Gaussian factors, heterogeneous mixtures, deep neural network factors, and time-inhomogeneous trellis factors, arbitrary order embedded Markov chains, a GUI graph viewer, and much more.
MLPACK is a C++ machine learning library with an emphasis on scalability, speed, and ease-of-use. Its aim is to make machine learning possible for novice users by means of a simple, consistent API, while simultaneously exploiting C++ language features to provide maximum performance and maximum flexibility for expert users. It contains algorithms such as k-means, Gaussian mixture models, hidden Markov models, density estimation trees, kernel PCA, locality-sensitive hashing, sparse coding, linear regression and least-angle regression.
fuzzylite is a fuzzy logic control library. Its goal is to allow you to easily create fuzzy logic controllers in a few steps utilizing object-oriented programming without requiring any third-party libraries. qtfuzzylite is a Qt-based GUI for fuzzylite. Its goal is to allow you to visually design your fuzzylite controllers and interact with them in real time.
QSMM, the "QSMM State Machine Model", is a framework for development of non-deterministic intelligent state models and systems with spur-driven behavior. It includes low-level functions for generating optimal actions by the system and high-level functions for building multinode models. In a multinode model, nodes represent components of a system you develop which choose optimal actions using the framework and can correspond to entities external to the system and which behavior is to be learnt. A node can choose optimal actions based on a current node state which is either set manually by your program or is identified automatically by the framework. Probability profiles for a state transition matrix and an action emission matrix of the node can be specified using an assembler program with a user-defined instruction set.
FreeFuzzyTime is a time reasoner based on Fuzzy Temporal Constraint Networks (FTCN), which treats fuzzy temporal information efficiently. It can be integrated into applications for diagnosis. This is especially important in areas like Intensive Care Units, where patients' data are handled by a temporal database. FuzzyTime uses a structure which consists of three levels of abstraction. The upper layer is the user interface, where a translator transforms the expressions introduced by the user into temporal relations between temporal entities (points and intervals). The semantics of a user’s expressions are analyzed and stored in the intermediate layer, or temporal world. Finally, the bottom layer is based on the FTCN model.
Wintermute is an intelligent framework of applications and libraries that uses neural networking to learn about its host. A pseudo-langauge engine that permits translations and grammar rulesets of any language to be incorporated into the system, and database downloads of different sets of data combine to provide a virtual self-thinking assistant that can be used to perform tasks like dictation to a text editor, and more complex tasks such as sorting of documents depending on the time of day, or automation of other routine tasks. It should be noted that Wintermute itself is a meta-project. It encompasses a large array of currently existing and potential produced projects.