Emdros is a corpus query system for storing and searching linguistically annotated text. It is very generic, supporting almost any kind of annotation from almost any linguistic theory. All linguistic levels of analysis are supported, including phonology, morphology, the lexical level, syntax, and discourse. The core libraries act as a middleware layer between a client and an underlying SQL database. MySQL, PostgreSQL, and SQLite are supported.
Glossword is a system to publish dictionaries, glossaries, and encyclopedias. It features an installation wizard, support for multiple languages, visual themes, multi-domain installation, an administrative interface with multi-user support, built-in search and cache engines, the ability to export/import dictionaries in XML format, and W3C-validated code. Glossword is useful for any sort of dictionary-like content, including sites with game cheat codes, online translators, references, and various kinds of CMS solutions.
queXC is a Web-based data cleaning and coding/classification system that takes a data file (such as data collected from a questionnaire) and cleans the text input fields by spacing them and spell checking them. It allows operators to code text fields to existing coding schemes, or to create a coding scheme on the fly. Multiple operators can code and clean simultaneously, with the ability to assign operators to do particular codes. The queXC system includes some coding schemes created from ABS (Australian Bureau of Statistics) data. It can be used as an open source replacement for Nvivo in some situations.
dbacl is a digramic Bayesian text classifier. Given some text, it calculates the posterior probabilities that the input resembles one of any number of previously learned document collections. It can be used to sort incoming email into arbitrary categories such as spam, work, and play, or simply to distinguish an English text from a French text. It fully supports international character sets, and uses sophisticated statistical models based on the Maximum Entropy Principle.
OpenEphyra is a question answering (QA) system. It retrieves answers to natural language questions from the Web and other sources. OpenEphyra comes with implementations of algorithms that proved effective in Carnegie Mellon's Ephyra system, which participated in the TREC evaluations. It is platform independent and can be set up in just a few minutes. The goal of this project is to give researchers the opportunity to develop new QA techniques without worrying about the end-to-end system.
SCAN is a personal information retrieval framework, combining search, text analysis, tagging, and metadata functions for document collections management. SCAN is a component-based software using a number of plugins for specific features. The basic SCAN platform can be easily extended with plugins for different document formats and document location types.
Dowser is a Web research and archiving tool that clusters results from search engines, associates words that appear in previous searches, and keeps a local cache of all the results you click on in a searchable database along with summaries and links to related information. It helps you to keep track of what you find, with no advertising.
Ellogon is a multi-lingual, cross-platform, general-purpose language engineering environment, developed in order to aid both researchers who are doing research in computational linguistics, as well as companies who produce and deliver language engineering systems. As a language engineering platform, it offers an extensive set of facilities, including tools for processing and visualising textual/HTML/XML data and associated linguistic information, support for lexical resources (like creating and embedding lexicons), tools for creating annotated corpora, accessing databases, comparing annotated data, or transforming linguistic information into vectors for use with various machine learning algorithms.
HALoGEN is an extremely powerful and easy to use general-purpose natural language generation system. It consists of a symbolic generator, a forest ranker, and some sample inputs. The symbolic generator includes the Sensus Ontology dictionary based on WordNet. The forest ranker includes a 250 million word ngram language model (unigram, bigram, and trigram) trained on the Wall Street Journal newspaper text. The symbolic generator is written in LISP and requires a Lisp interpreter.