InfoProfiler ^beta - Features

Information sources:

InfoProfiler can deal with all content sources where text is available in a machine-readable format e.g., World Wide Web, SEC Filings, Proprietary databases, and company Intranets.

Extracts Text and Information pieces:

InfoProfiler has inbuilt text extraction capabilities and can extract information from a website as well as from a given set of websites. It understands the structure of a webpage and decides whether it is a news, forum, review, bulletin board or a blog. InfoProfiler can also remove unwanted text portions (such as, advertisements) from a given information source.

Extracts Themes Concepts and Ideas:

InfoProfiler understands core concepts contained in a document or across a document corpus. This unique capability allows analysts to zoom in on the key ideas to derive insights from information.

Extracts Entities:

InfoProfiler understands entities like person names, roles, organizations, locations, biographical info, email, phone nos, and addresses. It can be configured to take user defined domain specific entities e.g. names of products, technologies, diseases, molecules etc.

Discovers Key Themes and Patterns:

InfoProfiler discovers relationships amongst key concepts and key entities - and allows users to compare and contrast pieces of information to gain optimum insights.

Generates Information Profiles:

InfoProfiler creates a summary of the document in the form of a semi structured intermediate report. It represents the concepts and themes from the document corpus as well as entities related to these themes. This information profile can be used as an input for creation of an insightful report.

User Input:

User input could be provided as a document corpus or a set of URLs or a set of keywords. The processed output can be searched based on key terms, themes or the entities that emerge from the documents.

System output:

The output could be presented to the user either in a user-defined format or as an information profile - in the form of a semi structured report.