Latent Dirichlet Allocation (LDA) is a tool for finding latent themes in collections of text documents. Formally called a Topic Model, this algorithm is similar to (and in some ways more robust than) earlier topic models such as Latent Semantic Analysis (Deerwester 1990) and Probabilistic LSA (Hofm...

plda is a parallel C++ implementation of Latent Dirichlet Allocation (LDA) (1). We are expecting to present a highly optimized parallel implemention of the Gibbs sampling algorithm for the training/inference of LDA (2). The carefully designed architecture is expected to support extensions of this ...

Large-scale, powerful and battery included! HadoopLDA can train LDA model with large corpus in parallel on a Hadoop cluster. It use distributed Gibbs Sampling technique, with built-in vocabulary selection. HadoopLDA is easy to use, a single command can turn huge amount of documents into a compact...

This project is an extension of the project csuFaceIdEval v.5 originally developed by Colorado State University. If you have any questions regarding the original project, please refer to following link http://www.cs.colostate.edu/evalfacerec/ The original project implements four base algorithms...

====Oct 12th, 2009==== LDA via gibbs sampling code is available now, you can download it from downloads page, file name:ldaviagibbs.zip. To compile and run, please refer to the HOWTO article in Wiki page. If you find any bugs, please tell me. ====Jul 31, 2009==== LDA Code & Latent Dirichlet Alloc...

We present the design and implementation of GLDA, a high-performance Latent Dirichlet Allocation (LDA) library running on graphics processing units (GPUs). LDA is an effective topic model used for many data mining and machine learning applications, e.g., classification, feature selection, and in...

Python implementation of the Latent Dirichlet Allocation(LDA). The learning algorithm is based on the Gibbs sampling.

The Hidden Topic Markov Model We propose modeling the topics of words in the document as a Markov chain. Specifically, we assume that all words in the same sentence have the same topic, and successive sentences are more likely to have the same topics. Since the topics are hidden, this leads to u...

Face Recognition Lab (facereclab) is a platform to implement and evaluate face recognition algorithms. The FaceRecLab is inspired by the csuFaceIdEval made at Colorado State University and is created in MATLAB for the purpose of ease of understanding and extending.

Topic models are statistical models of text. This package implements a number of topic models based on latent Dirichlet allocation (LDA).