NLP is a form of human-to-computer interaction where the elements of human language, be it spoken or written, are formalized so that a computer can perform value-adding tasks based on that interaction. Autonomy's approach differs from standard NLP use in that it is still able to harness the power of IDOL's conceptual analysis.
Autonomy's NLP technology functions independently of linguistic restraints, giving Autonomy's software universal application possibilities anywhere in the world.
Key Benefits:
Automatically leverages human interaction
Dynamic learning ability
Understands slang, industry specific jargon, sentiment and topics
Autonomy engages the users in many different conversation paths to help them disambiguate their result sets. By clustering the results according to concepts and parameters, and accessing the users' transactional and profile information to personalize the categorization, search results are meaningfully organized for quick filtering and decision making. Users can look to community insight to help find information, as well as rely on Autonomy's conceptual analytics to process the search results and continue the dialogue. In addition to offering all legacy tools like "Did you mean...?" to correct spelling errors, providing query suggestions as the user is typing into the search box, and offering facets according to content source type, Autonomy enhances the search experience using features such as:
Autonomy automatically produces a brief summary of each piece of content that is returned for a query. It generates three different types of summaries:
The conceptual summary displays a few sentences from the document that contain the most salient concepts (these sentences can be from different parts of the result document)
The contextual summary relates to the context of the original query - allowing the most applicable, dynamic summary to be provided in the results of a given query
The simple summary comprises a few sentences of the result documents
Instead of merely returning those users whose profiles indicate knowledge in the subject that was queried, AQCG returns and automatically clusters into communities those users who are most relevant to the queried person. For instance, if a user queried the name of the company's VP of Sales and he/she had been in frequent communication with client Z's finance and sales organizations at the time, two of the clusters that may appear as a result of the query would be employees in Z's finance department and the employees in Z's sales department, automatically organized into their respective communities.
Users can modify the relevancy algorithm used for their search at the time of query without any IT help
IDOL can incorporate all of the users' aforementioned enterprise 2.0 activities (i.e. social tagging, bookmarking) and users' content consumption behavior to improve and personalize relevancy. Expert behavior will have greater influence on the calculation by default. Documents they author, rank highly or have consumed recently will be positively affected. Since IDOL can monitor user's desktop activity, it can form user profiles of their current interests and expertise with great precision. IDOL can also adapt ranking of results so that documents that are consumed with great frequency following a certain query will be placed higher for subsequent queries relating to that same topic. This ranking can be adapted based on the behavior of the user, group or the entire enterprise community.
Business users and administrators are empowered to adjust the influence that each of these activities has on the relevance calculation, but users themselves can also individually adjust the influence of these factors per query. They can choose to search and rank documents based purely on IDOL's conceptual relevance calculation, on what the community deems relevant, on their implicit and explicit profiles, or some combination of the three.
IDOL automatically extracts entities for parametric filtering and navigation. A large set of entities are supported out-of-the-box, including SSN, credit card numbers, ticker symbols, places and names. When a user performs a query, results can be organized in a parametric hierarchy for guided navigation (parameters displayed to the user are adjustable by the business user and the administrator using the Autonomy Business Console). The parameters can also be derived from databases and other structured information, which in turn can be automatically merged with facets from unstructured content.
More importantly, users can easily merge multiple facets together to adjust their search criteria via drag-and-drop. There is no administrative involvement required to re-order, combine or remove facets during a search. For instance, if a search for "global warming" returned person facets (including Al Gore) and place facets (including Kyoto), users can simply drag the Al Gore parameter into the Kyoto parameter so that the newly formed facet of Al Gore + Kyoto will include documents that contain both parameters. In this way, users can determine the navigation hierarchy for ease of search.
By removing the administrative overhead of maintaining a hierarchy, organizations not only eliminate the costly exercise of developing a taxonomy, but also increase the likelihood of adoption since the hierarchy is dictated by the users themselves.
Following a query, IDOL displays all the prominent concepts extracted from each document in the result set. These "ideas" are derived from analyzing the entire text of each document and are not dependent on subjective metatags. The larger and bolder "ideas" indicate the appearance of that concept in many documents in the results list. The user can use the ideas cloud to focus his/her search even further. Clicking on an idea will refresh the results list with a more specific context that contains that concept.
In their respective profiles, users can define concepts, terms, rules, etc. that represent information of their interest. These entities will get consistently highlighted in the retrieval result set, together with the default IDOL highlighting for the executed query.
IDOL can certainly suggest queries for the user if they misspell the query text. However, IDOL goes beyond correcting mistyping and provides query guidance for correctly spelled queries that still return zero results. IDOL suggests similar, yet alternate concepts they can query, which are organized in a hierarchic cluster.
The Carousel Search user interface allows users to easily scroll through the results and quickly preview each content to gauge their relevancy. Moreover, the AQG function underneath the search query allows users to further refine their search results. The user has the option of clicking on the Open button at the bottom center of the window to also display search results in a conventional list alongside the carousel.
The Carousel Search user interface
This UI screenshot shows many of the powerful IDOL features interfaced following a search, including automatic hyperlinking, community feedback, ideas cloud, entity extraction and cluster tree
Forthcoming Events for Natural Language Processing (NLP)
Archived Events for Natural Language Processing (NLP)
Summary: ...Microsearch Search Service (Autonomy, Google, Baynote) Online Consumer Online Business Customer Centre Call Centre Enterprise Search Single Search Platform Capabilities Social Search Faceted Navigation Clustering Administration Security Connectors and Indexing Service Hyperlinking Scalability & Distribution...
Summary: ...results into clusters thus giving KLA employees a bird's eye view of data in the organization and allowing them to spot hot trends and topics • Automatic Hyperlinking feature suggests pertinent documents that are contextually linked to the original text thus ensuring that professionals have complete...
Summary: ...traditional approaches which depend heavily on grammatical structure or specific vocabulary. • IDOL's unique AQG refines search results into clusters thus giving KLA employees a bird's eye view of data in the organization and allowing them to spot hot trends and topics. • Automatic Hyperlinking feature...
Summary: ...Radioplayer - Autonomy Case Study. IDOL could be used to power search based on concepts, as opposed to keywords or the successive selection of metadata parameters (parametric search), or to analyze search results by grouping and interlinking them into clusters based on the similarities of their ideas,...
Summary: ...lists to narrow searches to specific categories such as “Location,” “Industry,” or “Date.” Parametric selection helps Linklaters more easily identify the correct organizations involved in every new deal. In addition, with the robust IDOL K2 Query Language working transparently behind the scenes,...
Summary: ...the relationships between content and client requirements on the basis of processes such as hyperlinking, clustering and categorisation. Hyperlinking suggests content that is contextually linked thus ensuring clients have full visibility over related content; information clustering provides a bird's eye...
Summary: ...IDOL K2’s easily modified business rules ensure all new and updated editorial content accurately populates the logical topic areas. In InfoMedia’s case, articles are hierarchically organized into a taxonomy with four top-level categories—subject, geography, company, and personal names—and as many...
Summary: ...Portal’s ACM Guide.” Adding Value to Premium IT Research Organization Association for Computing Machinery Industry IT Research Type of Application Subscription-based Library Autonomy Solution IDOL K2 infrastructure Functionality Parametric selection Conceptual retrieval High scalability Topics Language...
Summary: ...LLP (UK) to implement conceptual retrieval that transparently put the power of sophisticated queries behind simple, one-word searches. KPMG LLP (UK) is also well on its way toward implementing IDOL K2’s parametric selection. Using a combination of structured information, metadata, and free text, parametric...
Summary: ...Says O’Hara, “Interwoven provides us with a single, centrally-managed, matter fle to store pertinent e-mails and documents.” Desktop integration via FileSite allows the frm to drag and drop e-mails and attachments directly from Outlook into the relevant matter fle within WorkSite, for easy search...
Summary: ...with pre-defined folders for trading documentation, financials, due diligence reports, correspondence, and other content. Items dragged and dropped into a folder are automatically assigned the appropriate metadata, aiding search while eliminating the inefficiency and errors of manual profiling. SPECTRA...
This is a small selection of the Autonomy case studies available, please visit our publications site at http://publications.autonomy.com/ for more information.
Summary: ...spectrograph is able to display hot and breaking news in the same instance. The importance of clusters over time can be seen through the change of color and width. The color/ intensity of the lines is an indication of the size of cluster. The brighter colors indicate, “what’s hot” whilst the width...
Summary: ...query and the time of query. IDOL Retina provides automatic hyperlinking and automatically generated summaries A search for “global warming” will help users hone in on a more specific topic Parametric searching for items by their characteristics Autonomy Systems Ltd. Advanced Analytics Automatic Categorization...
Summary: ...Autonomy IDOL Server 7. Automatic Clustering Organizations can analyze large sets of documents and even user profiles and automatically identify inherent themes or information clusters. IDOL even clusters the unstructured content exchanged in email, telephone conversations and instant messages. IDOL uses...
Summary: ...data analytic technology Autonomy Virage identifies patterns in disparate data across multiple feeds to alert security officers to suspicious activities. A range of advanced functionality, including conceptual retrieval, dynamic hyperlinking, clustering, profiling and alerting, give instant visibility...
Summary: ...iManage Universal Search. Traditional search engines also fail to filter out documents that use the same words, but with an entirely different meaning than the user’s query. Simple keyword search techniques and relevance algorithms are unable to return quality search results, particularly when legacy...
Summary: ...iManage Universal Search. Traditional search engines also fail to filter out documents that use the same words, but with an entirely different meaning than the user’s query. Simple keyword search techniques and relevance algorithms are unable to return quality search results, particularly when legacy...
Summary: ...can be retrieved using event, time or date parameters. Bookmarks – bookmarks can be created manually or automatically to locate more easily important events. Drag and Drop Customization – Allows users to change camera views effortlessly. Exporting – To printer, external driver, or DVD for archiving...
Summary: ...and Configuration Multi-level default roles and configurable system access Configurable security options, including LDAP support, for integration into existing systems Clustering for high-availability XSL template and rendering language J2EE server architecture MySQL®, Oracle®, IBM DB2® and
...
Summary: ...data. This empowers the conflicts process with the richness of concepts and context, providing capabilities for query guidance, the extraction of names of companies and people from unstructured text, and the automatic categorization of search results into intelligent clusters, ranked by relevance. Product...
Summary: ...and edit the properties of a document or folder. The following default properties can be displayed: • Customising results You can change the way in which the Document Management Server displays responses by writing your own template (action template) for an Autonomy operation. • Document Retrieval...
This is a small selection of the Autonomy Product Briefs available, please visit our publications site at http://publications.autonomy.com/ for more information.
Summary: ...a new search result based on that idea. Eduction IDOL automatically extracts entities for parametric filtering. When the search result is displayed, users can easily drag-and-drop different parameters to combine facets, remove facets, or re-order facets. No administrative involvement is required. The...
Summary: ...Autonomy User Experience White Paper. • AQG (Automated Query Guidance) – helps the browser through his or her query by conceptually and contextually clustering the results along the way • Directed Parametric Navigation – the search through catalogs or databases is made effectively hassle-free...
Summary: ...without the user being needlessly diverted from their work in progress to perform a search or retrieval operation ? Automatic hyperlinking Completely removing the requirement to manually insert hyperlinks into content, Autonomy IDOL generates hyperlinks in real-time to all types of data, ensuring they...
Summary: ...with variable font size and boldness to represent the number of results that include that concept • Cluster tree: Displays hierarchy of relevant entities (concepts or metadata) extracted from the results list; it displays the count of documents that contain that entity • Automatic summarization of...
Summary: ...the text of a document or a court filing. These extracted entities can be used as additional query filters or to quickly find other documents that use the same citation or refer to the same judge. Also once extracted, users can setup email alerts that notify them of any event related to that entity.
...
Summary: ...itself. ? Clustering Autonomy's automatic clustering capabilities can take large sets of document data or even user-profile information and automatically identify the main set of information clusters (themes) inherent within your information assets. 1.2.3 Automatic personalization ? Automatic profiling...
Summary: ...Autonomy Whitepaper - The Power of Autonomy IDOL. Hyperlinking suggests content that is contextually linked thus ensuring clients have full visibility over related content; information clustering provides a bird’s eye view of data grouping similar items together to spot new or hot trends and topics;...
Summary: ...parameters (including multiple parameters) based on pre-defined rules e.g. void plus cashback An intuitive, user-friendly interface with drag and drop features makes the system easy to use and means minimal training is required. Users can search and view records based on transaction data and where CCTV...
Summary: ...content formats, 400 repositories, and more than 150 languages. Over 500 operations can be performed on digital content by IDOL, including hyperlinking, agents, summarization, taxonomy generation, clustering, eduction, profiling, alerting and retrieval. 3 “The Truth About Computer Assisted Review”:...
Summary: ...meaning enable advanced capabilities through the conceptual and contextual understanding of information. Some of the key functionalities of meaning-based platforms such as automatic hyperlinking and clustering are not available in traditional keyword search engines. For example, automatic hyperlinking,...
Summary: ...of inconsequential information, and also to enable effective disposition policies to be enacted. Another emerging requirement involves advanced search and navigation techniques such as concept searching, information clustering, dynamic hyperlinking and automatic query creation, all of which greatly increase...
Summary: ...E-mail Management: The New Imperative for Law Firms and Corporate Legal Departments. Ironically, although e-mail serves as the default collaborative platform for many companies, it can actually inhibit collaboration. Designed for one-to-one communication, e-mail used for many-to-many collaborative discussions...
This is a small selection of the Autonomy White Papers available, please visit our publications site at http://publications.autonomy.com/ for more information.
Related Case Studies
top of page
RSS Feeds
Find us on Facebook
Follow us on Twitter
See us on YouTube
Find us on LinkedIn
