Noesis automatically extracts meaning from a search phrase, in any language, and creates thousands of different ways to express that very same idea.
Overwhelming amounts of data, mostly textual, are produced every day. Even within a small office the amount of documents that accumulates over time is vast.
Making data classification and retrieval an even greater challenge are modern computer networks and the Internet, which make distribution and sharing of information inexpensive and virtually effortless.
We find ourselves flooded in data while unable to find specific meaningful bits of information. It"s like searching for a needle in a haystack.
To become valuable information, data needs to be analyzed and meaningful, in-context information extracted from it. Document management tools help save and categorize data. Still, finding specific information is difficult even when you know exactly what you are looking for.
Search and classification software use advanced Natural Language Processing (NLP) techniques to improve search results. But NLP does not really "understand" the ideas embedded in our textual query.
Search results produced by the most comprehensive search engines, using sophisticated NLP techniques and gigantic indexing mechanisms ? can not find exactly what we are looking for since the key to correct retrieval of information is the understanding of the underlying meaning.
BrainDamage's unique technology of automated acquisition of meaning, called Concept Identification (CID) enables search and retrieve applications that can truly find a needle in a haystack.
Noesis
- in Greek meaning understanding, perceiving, reasoning, learning. How does it Work?
Let"s look at the search query "financial fraud using id theft and off-shore banks".
Noesis automatically breaks down this phrase into meaningful "concept identification" such as "financial fraud", "id theft", "off-shore banks", etc. Each of these "finger-prints" leads to thousands of different ways humans verbally express the same meaning. After automatic review and analysis of its information sources, Noesis assembles the results most closely matching the meaning of the original search phrase.
By searching for meaning, Noesis is able to locate information by its meaning unlike other search engines that use word or phrase matching techniques to locate information.
How does Noesis help me?
Data repository classification (automatic, semi-automatic or manual classification of documents) Examples:
- Classification of news articles by specific subjects
- Classification of technical reports by specific fault criteria
- Classification of illness by symptoms
Information search and retrieval (based on local data repository, distributed, LAN based data repository, WEB and any combination) Examples:
- Search a legal documents repository for specific precedents and retrieve relevant documents
- Find all documents containing "2007 budget planning" information in a corporate document repository
- Identify all emails containing information pre-classified as confidential
Research assistant (finds information based on examples) Examples:
- Find cases with lab results similar to a given set of lab test results in a clinical data repository
- Automatically find alternative ways of spelling non-English names in English
Noesis Technology
- Is not Bayesian
- No morphology or stemming
- Does not use linguistic tags or rules
Noesis Technology is
- Language independent
- Purely empirical (text = experience)
- Fully automatic (only requires a large corpus of data to run)
Noesis System Components
Administrator Tools
- Server Module
- DB Maintenance Module
- Security and Authorizations Module
- Query Client
- Classification Client
CONTACT US
Email: danny_fine@braindamage.co.il
Web: www.braindamage.co.il
Post: BrainDamage Ltd., Pal Yam 2 (Oren Bldg.), Haifa, Israel, 33095