| Technology |
|
A Different Approach | | | Unique Technologies | | | Manual or Automatic |
|
A Unique Combination of Technologies
Over 130 Patents
Open Philosophy
Autonomy maintains an open philosophy with regards to the techniques it uses and is dedicated to selecting methods which optimize its technology, whether they are old or new. Accordingly, Autonomy embraces traditional or legacy methods such as keyword, Boolean, parametric and others. However, it is perhaps best known for its pioneering work in conceptual search based on computational pattern recognition (non-linear adaptive digital signal processing) and contextual linguistic analysis.
Built upon the seminal mathematical works of Thomas Bayes and Claude Shannon, and on a range of innovations that are covered by 130 patents, Autonomy technology identifies the patterns that naturally occur in text, voice or video files based on the usage and frequency of terms that correspond to specific concepts. By studying the preponderance of one pattern over another, Autonomy's technology understands that there is X% probability that the content in question deals with a specific subject. In this way, Autonomy extracts the content's digital essence, encodes the unique "signature" of the concepts, and enables a host of operations to be automatically performed on emails, phone conversations, video, documents and even people's interests.
Bayesian Inference
Thomas Bayes was an 18th century English cleric whose work has become a central tenet of modern statistical probability modeling. Bayes' efforts centered on calculating the probabilistic relationships between multiple variables and when new information is obtained, determining the extent to which these relationships are affected.
A traditional statistical argument posits that if a coin is tossed 100 times and comes up heads every time, it still has an even chance of coming up tails on the next throw. An alternative, Bayesian approach, is to say that 100 consecutive heads are evidence that the coin is biased. What Bayes theorem clearly demonstrated is that: a) the more information you are given, the more accurate your view of the world will be, and b) prior experience should be used to inform new data.In a typical problem such as judging the relevance of content to a given query, Bayesian theory dictates that this calculation be related to details that we already know.
Bayesian Inference
|
Shannon's Information Theory
|
A good example of this theory at work is Autonomy's agent profile technology. Users can create agents to automatically track the latest information related to their interests, and IDOL determines the relevance of a document based on the model of the agent. Adaptive Probabilistic Concept Modeling (APCM) algorithms are also used to analyze, sort and cross-reference unstructured information. In a similar manner, knowledge about the documents deemed relevant by a user to an agent's profile can be used in judging the relevance of future documents.
While most other models start with a prior knowledge of the state of the system and apply training to it, Autonomy begins with a blank slate and allows incoming data to dictate the model. In true Bayesian fashion, the model mixes new information with a growing body of older content to refine and retrain the engine.
Shannon's Information Theory
Shannon's Information Theory forms the mathematical foundation for all digital communications systems. Claude Shannon stated that information could be treated as a quantifiable value in communications.
Natural languages contain a high degree of redundancy or nonessential content. For example, a conversation in a noisy room can be understood even when some of the words cannot be heard, and the essence of a news article can be grasped simply by skimming over the text. Information Theory provides a framework for extracting the concepts from this redundancy.
Autonomy's approach to concept modeling relies on Shannon's theory that the less frequently a unit of communication occurs, the more information it conveys. Therefore, ideas, which are rarer within the context of a communication, tend to be more indicative of its meaning. It is this theory that enables Autonomy's software to determine the most important (or informative) concepts within a document.
Autonomy Technology White Paper| Technology |
|
A Different Approach | | | Unique Technologies | | | Manual or Automatic |
|
