My Roar in the News
Using natural language processing, MyRoar's technology is designed to provide financial services professionals (such as analysts) with a more accurate and efficient method of searching a variety of documents and sources
http://www.allbusiness.com/technology/software-services-applications-search-engines/13241755-1.html
If you're a financial analyst, a Boston startup called MyRoar Inc. says you can. The bootstrapped venture is building a semantic search product designed to answer analysts' questions directly, like, "Is anyone backing this bond?" or "How many hedge funds are there in the U.S.?".
http://www.masshightech.com/stories/2009/04/06/weekly30-MyRoar-brings-financial-analyst-search-results.html
"The system performed quite well on my test queries; for example, "What are the current financials for Parker Hannifin?" returned two results with the data I wanted. I will try to get Mr. Schiettecatte to participate in the Search Wizards Speak interview series. Give the system a whirl."
http://arnoldit.com/wordpress/2009/03/06/myroar-nlp-financial-information-centric-service/
This year we also want to mention a newcomer in search, MyRoar. Formed by technical folks who understand the financial services market, MyRoar is a natural language question machine. They claim that their technology does not need to be trained like vendors InQuira and Q-Go; we're hopeful to do more with them this year so we can report back on our findings.
http://www.ideaeng.com/tabId/98/itemId/181/2009-Overview-of-the-Enterprise-Search-Market.aspx
In today's current political and economic environment people have never had more questions. MyRoar helps people sort through the hype to find just the answers they are looking for.
http://www.altsearchengines.com/2009/03/09/the-question-answering-search-engine-myroar/
Using natural language processing, MyRoar's technology is designed to provide financial services professionals (such as analysts) with a more accurate and efficient method of searching a variety of documents and sources
http://www.allbusiness.com/technology/software-services-applications-search-engines/13241755-1.html
If you're a financial analyst, a Boston startup called MyRoar Inc. says you can. The bootstrapped venture is building a semantic search product designed to answer analysts' questions directly, like, "Is anyone backing this bond?" or "How many hedge funds are there in the U.S.?".
http://www.masshightech.com/stories/2009/04/06/weekly30-MyRoar-brings-financial-analyst-search-results.html
"The system performed quite well on my test queries; for example, "What are the current financials for Parker Hannifin?" returned two results with the data I wanted. I will try to get Mr. Schiettecatte to participate in the Search Wizards Speak interview series. Give the system a whirl."
http://arnoldit.com/wordpress/2009/03/06/myroar-nlp-financial-information-centric-service/
This year we also want to mention a newcomer in search, MyRoar. Formed by technical folks who understand the financial services market, MyRoar is a natural language question machine. They claim that their technology does not need to be trained like vendors InQuira and Q-Go; we're hopeful to do more with them this year so we can report back on our findings.
http://www.ideaeng.com/tabId/98/itemId/181/2009-Overview-of-the-Enterprise-Search-Market.aspx
In today's current political and economic environment people have never had more questions. MyRoar helps people sort through the hype to find just the answers they are looking for.
http://www.altsearchengines.com/2009/03/09/the-question-answering-search-engine-myroar/
How our Technology Works
Comparison to Key Word:
The MyRoar index is easy to implement and in its most basic form starts with a keyword combination search. Then pre-selected sentences are narrowed using computationally inexpensive natural language processing techniques. When the best sentences are selected they are sent to the question answering engine which runs in real time and returns the most accurate results.
Question Answering Technology:
People ask other people questions in natural language and they also want to search by asking questions in natural language. MyRoar takes user questions and converts them into language understandable and searchable by machines. The underlying technology relies on several machine learning components.
Why is the machine learning component so important? Traditional question answering software has included a “human” component, meaning someone is physically reviewing questions and searching for answers, aided by tools. The problem with this, besides the expense of paying for the “human” component is that people may not understand the variety of questions asked by professionals who are experts in their field, particularly when the “human component” is not likely well versed in their subject matter. The cost of having humans well versed in all subject matters for a semi-manual search would be prohibitive, resulting in substandard results. MyRoar’s machine learning, fully automated question answering system, solves this problem. Imagine a computer that can respond to mathematical equations better and faster than any human. That’s the power of question answering based on machine learning and natural language processing algorithms.
Questions are converted into statements. For example, “How much were Goldman’s earnings?” becomes “Goldman’s earnings were how much.”
Indexed sentences are stored recognizing each pronoun. Example: Goldman had a great quarter. They beat estimates by two times expectations earning $ 3billion in revenues. In this case “They” is replaced with “Goldman”, resulting in “Goldman beat estimates by two times expectations.”
Entities and Categorization. Here given the question, “How much were Goldman’s earnings?”, the excerpt “Goldman’s earnings were $3 billion.”, would be returned. A sentence that said, “Goldman’s earnings were interesting to most analysts.”, would not be returned. Also, given the question, “Who had good earnings?” the answer “Goldman had good earnings.” Would be returned, but not “A part of finance is to have good earnings.”
Synonyms. Given the example question, “Who had good earnings?”, both sentences “Goldman had good earnings.”, “Goldman had great earnings.”, and “Goldman’s earnings were pleasing.”, would all be returned. Syntactic Relationships. Given the question, “Did Goldman have good earning?s”, an answer that noted, “While Lehman had disappointingly low earnings, Goldman’s were good.”, would be returned while, “Lehman had good earnings, however Goldman’s were disappointingly low.”, would not be returned.
See how MyRoar Scales:
MyRoar’s design features offer a unique competitive advantage to traditional search by allowing users to ask real questions and get real answers while at the same time keeping costs low by using real time processing. The technology can scale to run in less than a second, even on an enterprise with terabytes of data! Another great feature of MyRoar is it runs on a separate index, which means no complex integration with current systems. This results in a time savings and cost savings upon installation.
Implementation:
MyRoar stores all of a companies enterprise documents in an index, keeping the appropriate permission settings. Documents come in a variety of formats, including but not limited to; PDF, Word, RDF, and HTML. The documents are preprocessed and stored in the index. Once indexed the documents run in real time, meaning only pre-selected answers are sent through the dependency matcher, resulting in an affordable installation for enterprises of all sizes.
