What Is The Dynamic Natural Language Engine?
  • The Dynamic Natural Language Engine is an advanced document management platform that utilizes innovative artificial intelligence (AI) technologies for text retrieval, organization, storing, searching, and manipulation. A key feature of DNLE is its ability to understand the meaning - or knowledge content - of a collection of words and sentences and build large-scale libraries of documents, from a variety of information sources, that are searchable by meaning and/or concept not just specific words and phrases.

  • DNLE utlizes proprietary concept-based ranking technology, enabling DNLE to find the documents that are conceptually most similar to a search phrase or document, not a long list of possible hits requiring manual sifting.

  • DNLE serves as the foundation tier for a wide array of possible applications, including: self-learning expert systems, automated call center applications, intelligent natural language search engines, security tools, smart devices, copyright protection systems, and applications for managing unstructured and semistructured data.

How Is DNLE Different From Search Engines?

The DNLE core technology is significantly different than current search engine technology in use today:

Although search engines (i.e. word indexing applications) can efficiently store the words in a document or web page to a database, and provide a mechanism to search for documents that have specific words, they don't understand the meaning of the words (and sentences) they are storing.

For example, a search engine can index the words: the quick brown fox jumped over the lazy dog, but it won't understand that the sentence is about animals. DNLE will index the words, but it also understands that the sentence is about two animals, that the animals can also be classified as mammals, that the sentence involves the motion of one animal moving over the other, and so on. This content signature is stored in the DNLE knowledgebase, and allows DNLE to search for and match documents based on meaning not just individual words. (i.e. DNLE will find a match when searching for: the agile dark-colored mammal leaped over the slothful vertebrate).

DNLE also has the uncanny ability to understand which words and phrases are most important in a document, and can give the important parts of a document more weight when searching for matches, thus enabling DNLE to provide the results that are most relevant. (or, even better, the single document that is best.)

Why Is DNLE Important?

Computers are not as useful as they could be because they are not very good at understanding the complexities and subtleties of human language.

Many times every day, search engines are utilized to answer questions, diagnose issues, and resolve problems. This typically involves many search iterations, using various combinations of keywords, and wading through pages of search results.

For example, the search phrase “my car is making an unusual sound” returns 30 million Google results. Researching an issue about your car using a typical search engine can be a very time consuming task with a low likelihood of success. A more useful search engine would monitor an automobile mechanic as he has conversations with his customers, storing the conceptual content of the conversations as two-part documents, a threaded context document and a response, allowing it to provide intelligent answers and - more importantly – engage in meaningful conversations with users, involving a question and answer process, ultimately leading to a better, more efficient, search experience.

DNLE is important because it provides the essential language engine that (1) understands the meaning of a collection of words and can efficiently store a content signature of that meaning, and (2) given a search document, can quickly find the single most similar document (based on meaning) in a large-scale document library.

  • The Dynamic Natural Language Engine is implemented as a Microsoft .NET Webservice.

  • DNLE has an integrated high-performance DBMS engine extended at all levels that utilizes proprietary data structures and algorithms that are extremely efficient.

  • DNLE can support large scale content libraries consisting of billions of documents while maintaining acceptable retrieval efficiency.

  • The development of DNLE has led to dozens of new ideas related to language theory and database architecture (we are actively pursuing patents).

Home | Contact | Terms | Privacy | Social | © Purple Inc.