MAIChem™ – Chemical Name Identification System

MAIChem is an identification system that gives researchers and information managers the means to automatically locate chemical names buried in collections of electronically stored data. It is an excellent processing system for patent document databases, technical report collections, scientific publication archives and similar data sets, to surface chemical names from the text.

What the software does

Finding chemical names in documents is a challenge. Because the number of potential chemical compounds is vast (unlimited, in theory) and there is a wide variety of ways a particular compound can be named, it’s impractical to merely match chemical names against unmodified text. MAIChem employs a multi-stage process.

MAIChem analyzes scientific text against regular expressions that match typical chemical morphemes such as “hydro” or “amine,” to see if the instances occur within words. After that, other algorithms are needed, for example, to distinguish between the non-chemical word “hydrophobia” and a legitimate chemical name “hydrogen sulfate.” MAIChem provides solutions for all of these challenges!

MAIChem interface (click here for a bigger view)

Powered by the M.A.I. Concept Extractor

This Data Harmony module deploys sophisticated matching algorithms based on both semantic and conceptual patterns, in combination with the MAIChem knowledge domain, to accomplish comprehensive identification of chemical names.

Feature – adjustable for the input data

In the dictionaries MAIChem brings to bear during comparative analysis of the input text, users can search for an item, add a new item or remove an item. Unique characteristics of an information collection can be reflected by these adjustments.

Implementation options

MAIChem is available for implementation as a standalone program or integrated in a network configuration for multiple-user access through Web browsers. Please contact your Data Harmony Technical Representative for more details.

Written by

Data Harmony