An application for visualizing a specific collection of authors, and the links or associations between them.  Links are determined by co-occurrence in the Orlando dataset.  The current dataset consists of authors, other people associated with them, and the text entries upon which these associations are based.  The application is designed in such a way that it could be extended to use other datasets as well.

The display can be customized in the several ways.  Node colours, shapes and sizes can be altered.  More significantly, the layout of the graph can be chosen from amongst several different algorithms.

As the collection is quite large, there are many filtering options available.  In terms of author, the following categories can be specified: gender, author/non-author, and author nationality.  The user can create complex search queries of the authors and the text, and specify a date restriction for the authors' life span.

Each association in the text has been tagged with one or more semantic attributes, i.e. the type or meaning of the relation between the people.  These tags have been organized into a tree, moving from general to more specific relations.  Each tag in the tree can be checked/unchecked, further filtering the results.

Finally, a data set and customizations can be saved as a session which can then be loaded at a later time.  In addition, screenshots of the current layout can be saved to an image file.

Scribe, now in version 3.5, is a free note-taking program available for both PC and Mac. Aimed particularly at historians, this program allows researchers to create digital note cards for managing sources, research notes, contacts, images, glossaries, outlines, timelines and more. Furthermore, it enables organization, keyword indexing, searching, linking and cross-referencing of all entries. Entries can be entered directly or imported from online catalogues such as Amazon.com. Information entered in Scribe can be exported directly to Zotero in addition to assembling, printing or copying formatted bibliographies directly from the program. Articles can be stored directly within the program, enabling researchers to search out and highlight particular words within the text in addition to the relevant researcher notes.

This prototype was produced as a proof-of-concept for the Criminal Intent project.  It allows the records of the Old Bailey Online project to be searched such that the returned results can be viewed as either a timeline or a concordance of terms.  The tool allows for reasonably complex queries to be placed and a range of different formats for the returned charts to be selected.  The tool is primarily built with Python and SQLite and makes use of the Google Chart API for some output options.

Mallet is a Java based library and command line framework that provides statistical and machine learning tools for use with natural language processing.

Collate was a program designed for scholars concerned with the difficulties of medieval vernacular traditions. It aimed to help scholars with the preparation of critical editions, and could collate up to a hundred texts. Collate was also capable of handling marked up text. It was available for Macintosh only, and could be purchased directly from its creator.

XTRACT was a tool for lexical collocation developed by Frank Smadja, then of Columbia University. It was designed to use statistical techniques to identify collocations of aribitrary length, and to generate syntactic relationships between words. This tool is no longer available and does not have a web presence.

TEXTPACK V is a historic collection of interrelated text analysis utilities first released for mainframe computers in the 1970s. With the fifth edition, released in the 1980s, it was ported from FORTRAN to run on PC.

INRAC was an early language compiler useful for prototyping natural language processing interfaces, designing conversational systems, and experimental prose and poetry.

Parser 1.0 was a commercial program running in DOS for analyzing English-language text, aimed at academics and other researchers.

FANGORN was a historically important programming language designed for linguistic analysis in the humanities.