Autonomy for Investigative

Autonomy's categorization capabilities remove the necessity for intelligence agencies to rely on human intervention. Autonomy's categorization features allow users to derive precise categories through concepts found within unstructured text. This also ensures that all data is classified in the correct context and with the utmost accuracy.

The information that Autonomy has automatically aggregated and categorized is presented to users in the form of channels. Channels facilitate the creation of a directory structure that is both easy to navigate and maintain. Users with appropriate administrative permission can create and administer these category channels for all other individuals in the organization.

This means data becomes more accessible to investigators, reducing defocusing processes and overheads in information management.

Cluster information is hierarchically agglomerated data that has been extracted from snapshots (this does not require the setup of an initial taxonomy). Each cluster represents a concept area that contains a set of items, which share common properties. Clustering data allows users to make trends and developments in data visible.

Through cluster visualization Autonomy simplifies analysis with two intuitive Java-based user interfaces.

Clustering and Cluster Visualization enables investigators to analyze large sets of data or even user profile information to automatically identify knowledge gaps that need to be addressed, reveal hidden areas of expertise that can be transferred and even recreate digital sequences that reveal the meaning and intent of individual and corporate behaviors, identifying themes, issues and connections.

IDOL allows content to be searched simultaneously in any language and any format, wherever it is stored, and presented to users with automatically generated summaries and hyperlinks to similar information and in real-time. Tools include Natural Language Retrieval, Query By Example, Refine By Example and Cross-Language Search.

In the context of investigation, conceptual retrieval dramatically improves access to information as regulators, examiners and investigators are given immediate access to the documents they need. Using Autonomy's concept-based technology, government analysts and investigators will be able to use natural language to describe what they are tracking or what they have heard. Because the system analyzes and understands the concepts within text, video and audio content, rather than relying on keywords, it is able to identify patterns or clusters of words that when combined together may signal code for dubious activities.

In addition, investigators need to gather and analyze vast amounts of information in the shortest amount of time. Often, examiners do not know the exact type of information they are looking for. Conceptual retrieval includes the fast analysis and identification of both 'known' and 'unknown' subjects, which allows investigators to be alerted to previously unidentified information.

Built on a unique pattern-recognition technology, Hyperlinking enables the manual or fully automated identification and matching of similar pieces of information in real time. The ability of Autonomy software to identify vital relationships between different pieces of information is a key advantage for investigators who save time having to manually navigate to related information. In addition it reduces duplication of effort and ensures key people remain informed and up-to-date.

Autonomy's Collaboration and Expertise Networks automatically identify those individuals and expert groups in organizations who have the greatest knowledge in any required field. These experts can be brought together to build communities of expertise, fuelling collaboration and discussion forums to realize common goals.

The investigation process can be hindered by natural manual boundaries such as disparate groups of investigators who are unaware of each other's activities and discoveries. In addition, there is the potential problem of information refraction caused by poor interpretation and conflicting priorities of various investigative groups who require access to the same corpus of information.

CEN facilitates communication between different expert groups by storing user profiles and employing them to identify and bring together highly focused experts. It can also assist a legal team who need to drive decisions by collaboratively sharing documents and themes of investigation.

In the context of the fight against terrorism, the U.S. Department of Homeland Security uses CEN as an advanced knowledge sharing solution for its agents, enabling the identification of experts knowledgeable about specific topics such as technology, geography, weapons and political regimes, within the government.

Alerting enables investigators to be instantly alerted to pertinent content, without the need for any manual input. This ensures that their attention is brought to key material that otherwise could have been missed completely. The Securities and Exchange Commission (SEC) uses this feature to identify indicators of potential problems rapidly, evaluate compliance issues and discover relevant evidence buried in millions of documents. This radically speeds up the investigation process. In addition, by automatically searching millions of email messages and attachments and alerting users to any relevant content available regulators, this feature enables examiners and investigators to stay on top of numerous pending cases at the same time. In the Department of Homeland Security, alerting is used to deliver intelligence to agents on new developments in a particular case.