Cog Blog

Cognition’s New Advanced Syntactic Parser, Combined with the World’s Largest Semantic Map of English, Enables Tag-Free Semantic Search

LOS ANGELES – November 5, 2009 — Cognition  Technologies, the next-generation Semantic Natural Language Processing (NLP) company, announces a significant advancement in its Semantic Search capabilities through the release of its advanced syntactic parser module.  Cognition’s powerful Search technology is currently available as an embedded component within applications to make them “semantically aware.”  Examples of this include LexisNexis Concordance and Merrill Lextranet within the legal e-discovery and litigation support markets, where it is being used to cut document review time and costs by 30-50 percent and significantly improve search precision and recall across large data sets. Other applications enhanced by Cognition’s advanced Semantic NLP include interpretation of voice interaction, user question interpretation, mobile Search, customer sentiment analysis, medical informatics and publishing.
Read the rest of this entry »

Cognition Technologies has recently added an advanced syntactic parser module to its language-understanding technology. “What does that mean?”, you ask? It means that Cognition can now “parse”, or break down, the component parts of sentences to deliver an even more accurate and complete understanding of the content. Since words often have more than one meaning, the ability to parse sentences enhances the technology’s ability to understand the context and sentence structure of the material being analyzed.

Cognition’s Advanced Syntactic Parser enhances Cognition’s Search technology by making fuller use of the meaning relationships between the words in a sentence (e.g. context), which is a key piece in the language-understanding puzzle. The parser improves upon the already high accuracy of our pre-parser system by significantly reducing error rates in word meaning disambiguation (i.e. understanding the correct meaning of a word based upon the context in which it is used).

“Cognition’s parser is also a significant stepping-stone to future capabilities which will take the technology well beyond just Search functionality,” said Dr. Kathleen Dahlgren, Cognition’s founder and CTO. “Examples include enhanced reasoning within and across sentences, anaphora resolution (i.e., identifying the referents of “he” or “them”, etc. within a document), and the development of a true question-answering system in the future. The parser is a critical component needed to bring a fuller sense of understanding to the content housed on the Web.”

Here are some examples of Search accuracy improvements due to the introduction of the new advanced parser are:

Query: “What happened after Rome fell to the Visigoths?”

The pre-parser system incorrectly interprets “fell” as the stem “fell” as in “to fell a tree.” The advanced parser recognizes that “fell”, followed by a prepositional phrase, is a better match for the “concede or defeat” sense of the stem “fall” and correctly picks that stem.

Query: When did the U.S. begin building monuments?”

The pre-parser system incorrectly analyzes “building” as the noun meaning “edifice or structure”. The advanced parser system recognizes that “building” must be a verb in the sentence and correctly picks the “construct” meaning of the stem “build”.

Read more info here, or try the advanced parser for yourself at www.Cognition.com. The Cognition Website enables you to semantically search Wikipedia, a database of Supreme Court and Appellate Court cases in the Federal Court system since 1950, the Medline medical abstract database, and the four Gospels of the New Testament.

The Merrill Corporation, a leading provider of outsourced solutions for complex business communication and information management, recently launched its Merrill-Lextranet™ version 5.6 which features an integration with Cognition Technologies’ Semantic Search and discovery technology.

Cognition helps Merrill-Lextranet with early identification and review of key evidence, and enables their e-discovery clients (i.e. law firms) to conduct more meaningful content searches and uncover relevant evidence more quickly and at a lower cost.

Cognition improves Merrill-Lextranet’s review process by:

• Reducing the number of documents needing manual review (i.e. culling out irrelevant and unrelated information);
• Finding all documents about a given case issue, regardless of how the issue is worded (i.e. Cognition enables the user to search on concepts, not just key words);
• Finding all documents in a richer set of documents with fewer unrelated or irrelevant documents, thereby saving significant Reviewer time and money;
• Forming conceptual Booleans that organize the documents into case-issue related folders which are more conducive to review because all documents in the folder are about the same issue, and because there are fewer documents to review;

For more information on Cognition Legal, please click here.

SearchEngineJournal.com wrote an article entitled “9 Semantic Search Engines That Will Change the World of Search” and included Cognition Technologies in the group.

Read the full article here.

In advance of the Boston Search Engine Meeting, Dr. Dahlgren, Cognition Technologies’ CTO and founder, was interviewed by Harry Collier of Infonortics and the interview was posted on Stephen Arnold’s ArnoldIT.com site.

It is a great interview and a cutting-edge look at the future of search as seen through the eyes of our very own Dr. Dahlgren.  In the article she discusses what she views as the three major challenges in the search field, as well as many other important facts about Semantic NLP and the Semantic Web.

Check out the full article here.

Great article about Merrill Lextranet, a Cognition Technologies’ customer:

After evaluating and using Merrill Lextranet’s 5.6 version of its case management solution, we would definitely place it in that category and consider it to be among the best-in-class solutions currently available.

Cognition Technologies is part of Merrill Lextranet’s solution:

Lextranet has augmented its native search capabilities with conceptual search, in this case by incorporating third-party software from Cognition. Conceptual search looks for documents not by matching keywords, but rather by identifying documents containing words related to a concept. For example, a conceptual search for the word “airplane” might return documents that do not contain the word “airplane,” but do contain the word “glider” or “helicopter.”

Read the full article: “EDD Case Management with Lextranet”.

In December I spoke on a panel at SES Chicago about how Semantic Search will change our lives. I wanted to do a blog post about what some of the speakers said.

Semantic Search means various things. To several of the speakers on this panel it means search based on recovery of the detailed linguistic meaning of the query and target document base. To others it means recovery of “sentiment” or evaluative language in customer reviews of vendors such as restaurants. To others it means search in semantically tagged fields or structured data in a relational database.

Several of the speakers in the panel agreed that a radical shift in search in which meaning-based indices are searched, rather than pattern-based indices, is going to make the next big improvement in search. Similarly, ads will be placed based on meaning. As Tim Musgrove of TextDigger said, it’s silly for various vendors with different types of businesses all to fight over the keyword “palm”. The word (or pattern) is relevant to palm trees, palm pilots, or palms of the hand (say for a sporting-goods vendor). If linguistic semantic search replaced string search, there would be three different keywords (for different meanings of “palm”).

The Powerset speaker (Scott Prevost) agreed. He explained that Powerset linguistic processing enables their search engine to distinguish “who did what to whom”. This avoids false hits due to argument structure, so a query about “Who did Merrill-Lynch acquire” does not retrieve a document about Bank of American acquiring Merrill-Lynch. Also, Powerset’s linguistic semantics enables their search engine to find facts about query entities in free text, and use those facts to enhance search result displays.

Larry Cornett of Yahoo! described the use of structured data along with free text to improve the effectiveness of retrieval results. Developers are allowed to add structured data and create applications that modify and augment the display on the Yahoo! results page. For example, a restaurant review developer could show ratings, reviews, address, and other information most users want about a restaurant, directly on the results page. This type of semantics dovetails with Web 3.0 tagging in that structured semantic information is introduced by hand through knowledge engineering.

Nadaraju Bandaru of Boorah! explained “sentiment extraction”, where reviews of offerings such as restaurants are rated on sentiment scales to decide whether they are “boo” or “rah” ratings. These results are displayed to help users quickly decide upon their choice. Sentiment scales are determined by the semantics of words in the reviews such as “yummy”, “delicious”, “nice”, etc. Boorah! also illustrates the advantage of focusing on a vertical to narrow the range of meanings, improving precision.

This panel suggests that there is significant interest in improving the search experience with semantics, especially deep linguistic semantics a la Cognition and Powerset.

We are excited to announce that we have put up a new demo site that allows users to search the Gospels of the Bible (Matthew, Mark, Luke and John). 

http://Gospels.Cognition.com  or from our Home Page at www.Cognition.com.  

As you may know, Biblical text varies by the translation used,  and the Bible uses extensive metaphorical language to express concepts.   However, even with all of these idiosyncrasies, we think Cognition’s Semantic NLP does a pretty good job of understanding the content, as well as providing answers to questions and references for Biblical phrases and other queries.

We have worked very hard to show companies interested in semantically-enabling their technologies that Cognition’s technology understands language and concept nuance.  Try out our latest demonstration utilizing the Gospels, and let us know what you think.  Click on a few sample queries or do your own, then check out the results.  Be sure to look at the “Meaning Box” on the right side of the results page to see how well we understood what you were searching for.

From all of us at Cognition, have a great Holiday Season and a fantastic New Year!

The latest Pew Internet and American Life report entitled “The Future of the Internet III” asked “Internet leaders, activists and analysts” a variety of questions.  One finding that we in the Semantic technology world should pay attention to was “Network engineering research will build on the status quo—there isn’t likely to be a ‘next-gen’ Internet”. 

The report notes:

Nearly four out of five respondents (78%) said they think the original Internet

architecture will still be in place in 2020 even as it is continually

being refined. They did not believe the current Internet will be

replaced by a completely new “next-generation” system between

now and 2020. Those who wrote extended elaborations to their

answers projected the expectation that IPv6 and the Semantic Web

will be vital elements in the continuing development of the Internet

over the next decade.

 See the full Pew report here.

I read David Provost’s report from September 2008 entitled “On The Cusp: A Global Review of the Semantic Web Industry” and it got me re-thinking how Semantic companies gain adoption through the creation of applications utilizing their Semantic technologies.

It seems to me that Semantic companies have to do one of two things to create value in an application:

1) Innovate a completely new application.  This puts Semantic companies in competition with every single Web/Software company and entrepreneur out there, not just the other Semantic companies.  Plus, companies have to compete against the noise of tens of thousands of Web apps already out there.

2) Improve upon existing apps that do not use Semantics (and show such an improvement that users can immediately see a difference).  I would argue that here Twine is going to face an uphill battle (and I like Twine!).  They’re a social network/”interest” application that is directly competing against many large social networks and Web discovery companies (eg Digg).  But to be successful, they must compete for, and win over, a technology agnostic audience.  This audience couldn’t care less about Semantics for Semantics’ sake, and Twine ends up competing against Facebook’s feature set and already huge network effect.

Both of these approaches are challenging.

What Semantic applications would you like to see?