• History
• Management
• Cogito – a unique technology
• Why Semantic Intelligence?
• Awards

Click here for more Awards

Why Semantic Intelligence?

More than 80% of the knowledge on which our daily jobs are based is unstructured and available in textual form. Intellectual Capital (know how, patent, brands, rights, projects) and web resources are the real center of our daily activities: every day we are buried by emails, documents, web site pages, articles, researches, reports... a huge amount of text which contains potentially relevant data.

Instead of being a great opportunity for growth and development, this is becoming the biggest hurdle in defining strategies, in the activities of analysis and control, in managing a project.

Information is nothing if it's not managed.

Having access to documents without knowing of their content is like not having access to this material at all.

It is not possible to manage the knowledge we have access to, counting only on manual access, because we don't have enough time. There are too many things to read, analyze, summarize and compare. It is not easy to find what we need at the time when we need it, and then process it, integrate it with other relevant content, and finally share it with our coworkers.

In knowledge management, technology is a must.

The decision is about what technology to choose.

There are three kinds of technology to manage unstructured information:

• Keyword based, with the possibility to add statistical elements

• Shallow linguistic

• Deep Linguistic or semantic, that is, the only technology that is based on understanding the meaning of the content.

Keyword based technology

The text is divided into the list of words it contains without considering their meaning. The list is cleaned up excluding all the elements that have a high frequency (like articles, prepositions.) The text filtered with these criteria becomes just a list of strings of characters that appear a certain number of times.

Actually, it appears to us in the same way as when we read a document in a language we don't know.

For example, let's consider the following paragraph in Hungarian:

Az IBM szokásosan nagy hangsúlyt helyez a továbbképzésre, így munkatársai évente számos szakmai tanfolyamon vesznek részt. Az elmúlt években a csoport több tagja is részt vett több hónapos, egyesült államokbeli, angliai illetve németországi projekt munkákban, melyek során nemzetközi csoportban végeztek fejlesztoi tevékenységet.

If we don't know the language we cannot understand the meaning: nevertheless we can still understand that szokásosan, nagy, hangsúlyt, helyez… are words that probably have a meaning and it is worth it to include them in an index: the keyword index, that is. This means that in a search, the system would be able to find only the documents containing the words written exactly as the user writes them.

This means that a phrase like…

Last year we closed two plants in Michigan and this year we are opening two in China.

… will not be found using the most common natural search a user would do in this case: "closing plants".

Shallow Linguistic Technology

If the words contained in the text are recognized, they are automatically linked to their respective lemma. This means that in a search for the word "sell" the user will be able to retrieve all documents including the word "sold", "selling", "sell", "sells".

However, this means that for a shallow linguistic system, there is no difference between the following two texts:

text 1

New York, April 27- The Dow fell 46.58, or 0.42 percent, to 11,002.14. The Standard & Poor's 500 index fell 1.44, or 0.11 percent, to 1,263.85, and the Nasdaq composite gained 6.84, or 0.32 percent, to 2,162.78.

text 2

New York ,April 27- The Dow gained 46.58, or 0.42 percent, to 11,002.14. The Standard & Poor's 500 index fell 1.44, or 0.11 percent, to 1,263.85, and the Nasdaq composite fell 6.84, or 0.32 percent, to 2,162.78.

Even if the two texts are perceived as the same, in reality they are significantly different…

Semantic Technology

Semantic analysis includes grammatical, logical, morphological analysis and the identification of concepts. In other words, the different acceptions of terms are correctly identified…

for example, when the word "stock" is used as a synonym of "broth", "share of a company", "goods stored/available in a warehouse"

in addition, the different ways to express the same meaning are also identified…

for example "car" , intended as a vehicle running thanks to an engine, is the equivalent to "motorvehicle" or "sedan", "convertible", "SUV" or even "Ford", Toyota", "Mustang" etc.

In reality, the text is analyzed in a similar way in which people analyze.

From Knowledge Management to Semantic Intelligence

Semantic technology enables the management of complex issues and bypasses the "search" with all its limitations to move directly to "find, organize, select and correlate."

Similar to what Business Intelligence is in the world of structured data, the best solution for the management of unstructured data is Semantic Intelligence.

Expert System participates in the 2010 Forrester Groundswell Awards Business to Business Category, 9-2-2010

Upcoming Events - Autumn 2010 Join Expert System at These Upcoming Events!

Expert System Highlighted as Leader in Enterprise Semantic Software Gilbane Group Report Analyzes Evolving Landscape of Semantic Technologies, 8-24-2010