Home
Turning Data Into Actionable Assets

Glossary

Actionable Asset
An actionable asset refers to the mission critical data and knowledge that drives the decision-making process and guides business transactions.  Data that is unstructured, inconsistent and incomplete is of little value in supporting decision making. Unfortunately, only about 20% of business content is in a structured and useable format. Gleaning consistent, complete, actionable data from jargon-rich technical information is often a difficult, costly and labor intensive process.  XSB's automated data management solutions transform unstructured data into actionable data which is rich, structured, precise and timely. 

ACTRAD
ACTRAD is an acronym for XSB’s Acquire, Classify, Transform, Reason, Advise, Decide methodology. Based on this process, our tools enable business users to extract, integrate, and process vast amounts of unstructured “dirty” data in a real time environment
.

Agent
See Automated Agent

Automated Agent
An automated agent is a computer program that achieves a set of defined goals, or performs a defined action, independent from user control or human intervention. Some agents may have "intelligence" in the sense that they prioritize their goals and reason about how to achieve them efficiently in a changing environment. Certain agents may even collaborate or negotiate with one another to accomplish their goals.

Specifically, XSB, Inc.'s XRover® automated agents are web agents which retrieve data available on the World Wide Web. As this data is often poorly structured and organized, a great deal of manual labor would be needed to locate and extract relevant information. Web agents assists users in the process of performing precise searches to locate specific data of interest from the vast amounts of Web available data. Once located, the agent is able to extract this information and output results to a Microsoft Access or .XML file.

Attribute-Value Pairs
An Attribute-value pair refers to the name of a property for an object along with the value associated with that property. For instance (Price/$1.00) might be an attribute-value pair of a given part.

CDF

The Coherent Description Framework (CDF) is XSB’s proprietary ontology management system. Using CDF, large ontologies can be constructed indicating information about classes and their relations (objects and their attributes). CDF includes a visual editor for concurrent editing of an ontology, and a module-like component system, which makes it easy to combine ontologies from different sources into larger application-specific ontologies.

Coherent Data

Coherent data is information that is rich, precise, and timely.  It is the backbone of decision making; it is highly structured, readily queriable, and integrated across the entire enterprise.

Coherent View®
The Coherent View® refers to a structured and amalgamated view of data assembled from multiple semi-structured web, document, and legacy data sources. It presents objects from these sources along with their identifying attributes. It also preserves traceability of the presented information back to the original data source.

Data Classifier/Data Classification
See Ontology Directed Classifier

Data Mining
Data mining (or data acquisition) is the process of autonomously retrieving useful information or knowledge (“actionable assets”) from large data stores or sets. Data mining can be performed on a variety of data stores, including the World Wide Web, relational databases, transactional databases, internal legacy systems, pdf documents, and data warehouses. Many of XSB, Inc.’s solutions involve automated data mining or information retrieval.

Deep Web
The term Deep Web refers to information served up on Web sites that is hidden or generally inaccessible through traditional search methods. For example, information that resides or is provided through searchable databases, the results of which can only be discovered through queries or by filling out Web based forms is considered Deep Web data. XSB's XRover® Web agents are software robots that can locate and extract deep web data which is not accessible via traditional keyword-based search engines.

InterProlog
Interprolog provides a mechanism for the XSB programming language and the Java programming language to communicate. InterProlog translates XSB terms to Java objects and Java objects to XSB terms, so that Java can call XSB and XSB can call Java. Communication can be established either through the Java Native Interface, or through the use of Sockets. InterProlog was originally developed by XSB’s partner, Declarativa Inc. and is open-source in nature. InterProlog can be freely obtained through Declarativa's Web site located at www.declarativa.com. XSB, Inc. has funded its further development for the open-source community.

 

Master Data File

A Master Data File contains clean, structured, consistent data that can be instantly accessed and queried by any business unit within the organization.

NAICS
NAICS stands for the North American Industry Classification System.  The NAICS taxonomy, which recently replaced the SIC taxonomy, is the most extensively used classification system for businesses. XSB's Ontology Directed Classifier uses this taxonomy, among others, to classify part and service data. XSB uses this taxonomy, among others, to classify part and service data.

Ontology

An ontology is a defined knowledge representation format. That is, an ontology is a shared understanding of the structure of a domain of interest. Ontologies make it easy both for human beings (not necessarily computer programmers) to compile and maintain a body of knowledge, and for computer programs to use this knowledge to intelligently manipulate data.

An ontology organizes all data using the concepts of class, object, and relationship. Classes are organized into a hierarchy, ordered by subclass, called a taxonomy. A well-known taxonomy is the biological taxonomy of all living things, in which living things are subclassed into their kingdom: plant or animal. Plants and animals are further classified into phylum, etc. An ontology extends a taxonomy by including relationships among objects and classes, which can represent properties and values. To continue the biological example, there is a relationship "number of limbs" between certain classes of animals and integers.

Many taxonomies have been developed to organize knowledge in particular areas. For example, the UNSPSC taxonomy attempts to structure all parts and services,  the NAICS taxonomy organizes company capabilities, and the SNOMED taxonomy describes the medical domain. XSB's Ontology Directed Classifier uses these taxonomies, among others, to classify part and service data.

Ontology Directed Classifier (ODC)
Classifying data to a taxonomy can help categorize data into a form that improves search and analyses. Classifying data to an internationally supported taxonomy ensures that data is transferable across multiple supply chains. Automating the process of data classification to a target taxonomy can greatly improve data quality across the enterprise and reduce the considerable time and expense associated with implementing and maintaining a common coding schema. XSB's automated Ontology Directed Classifier (ODC) tool, is designed to classify data based on any public or proprietary taxonomy. Additionally, ODC enables users to easily map from one taxonomy to another. Pre-trained classifiers can be licensed with specific knowledge for both the UNSPSC and NAICS taxonomies.

Ontology Directed Extraction (ODE)
Much enterprise data is presented in free text or some other poorly structured format. In order to make this data readily usable, it must be standardized, and inference rules applied to resolve inconsistencies in the text and infer missing data. Using a powerful reasoning engine that mirrors human cognition, XSB's ODE software application extracts attribute-value pairs from textual descriptions of objects so that they can be classified and structured to a pre-defined or customized ontology.

Ontology Mapping
The process of ontology mapping concerns how classes from one ontology can be mapped to classes of another taxonomy in an automated way.

Product Equivalence
Many similarities exist between physical goods and services. Businesses users are often required to determine when two different products or goods are comparable - is a given HP toner cartridge the same as a given Canon toner cartridge? Are the two products interchangeable? Product Equivalence refers to the process by which a user determines whether or not two items are equivalent or are so similar that they can be used interchangeably. XSB's approach to determining Product Equivalence is as follows: two products are deemed equivalent if they have the same core attributes or properties. To determine this, the products are classified using the Ontology Directed Classifier, and information about the products is extracted from text and/or inferred by the Ontology Directed Extractor. These items can then be judged as equivalent or not based on a comparison of their key attributes.

Regular Expression (RE)
A template or pattern for a text string. A regular expression indicates, in general terms, what characteristics the text must have to fit its template. XSB's automated XRover® agents use regular expressions to characterize the type of data they wish to extract. Data fitting the pattern of a particular regular expression is matched and then extracted.

Society of Agents
A Society of Agents is a group of inter-related automated agents that work together and communicate with one another to accomplish a set of pre-defined goals.

Standardization/Data Standardization
Standardization is an important part of many data management processes. The need for Data Standardization arises because certain objects may have many different textual strings or names associated with them. For instance, the company International Business Machines may often be designated as IBM, or IBM, Inc, or Intl Bus. Mach, and so on. Erroneous or inconsistently represented data can lead to costly mistakes; therefore businesses must standardize all of these various representations into an single unified canonical structure. XSB, Inc. has a core competency in the standardization of company names and addresses.

Taxonomy
A taxonomy is a hierarchical organization for sets of objects. Taxonomies are used to group similar kinds of things together, placing more general definitions at the top of the hierarchy and more specific ones at lower nodes in the taxonomy. Two of the most well known and widely accepted global taxonomies are the United Nations Standard Products and Services Code® (UNSPSC) (http://www.unspsc.org/) and the North American Industrial Classification System (NAICS) (http://www.census.gov/epcd/www/naics.html) taxonomy. XSB's automated data classification solutions categorize data into a form that improves search and analyses. Classifying data to an internationally supported taxonomy ensures that data is transferable across multiple supply chains. 

UNSPSC
The United Nations Standard Products and Services Code (UNSPSC®) is a taxonomy of goods and services devised by the United Nations in order to facilitate electronic commerce. XSB's Ontology Directed Classifier tool uses this this taxonomy to extract information about products and classify them.

WEAVE®
WEAVE® is an automated Web based supply chain optimization tool that aggregates multiple on-line Web catalogs with the company’s own stock room inventory to produce an integrated virtual view of the supply chain. WEAVE® provides a Web based interactive search tool enabling users to locate and compare items across the data landscape so that similar items can be quickly identified and pricing from various suppliers can be easily referenced.

XJ
The XJ open source system allows the XSB system to serve as a presentation logic layer for graphical user interfaces. In an XJ program, certain XSB terms can be interpreted as models of Java Swing objects. XJ is built on top of InterProlog, which performs the actual mapping of the XSB terms to the Java objects. Graphical user events are captured by Java and depending on the XJ program, either handled directly in Java or passed to XSB to be handled. XJ has been developed in collaboration with our partner Declarativa Inc.

XRover® Agent Technology
XRover® is XSB's automated intelligent agent software which is designed to mine Web data that is not readily accessible through traditional search technologies. XRover® Web agents go to user specified Web sites, follow links, fill out forms, and precisely retrieve product information of interest to the user. In contrast to general-purpose Web crawlers, XRover® agents can retrieve information from dynamically generated pages. They are also able to bring back not just the contents of a Web page, but selectively retrieve product information contained at specific areas of Web pages. 

XRover® Agent Manager
The XRover® Agent Manager is XSB, Inc.'s desktop software application that provides users with the infrastructure needed to view and manage the execution of agent tasks.


XRover® Agent Validator
The XRover® Agent Validator is a tool that examines a completed agent sitemap in order to identify any errors or problems that may interfere with the proper functioning of the agent.


XSB Logic Programming System/XSB Prolog
XSB Prolog is a logic programming and deductive database system that was originally developed at the Computer Science Department of Stony Brook University. The XSB system is a highly optimized, re-engineered extension of the Prolog programming language, blending intelligent database technology and advanced logic programming. Most of the XSB system itself is open-source, and available at xsb.sourceforge.net. XSB, Inc. has developed many applications and extensions of XSB some of which are proprietary, and some of which have been donated to the open-source community.

Xtractica®
XSB's Xtractica® Product Content and Data Management solution provides a unique automated alternative to transform an enterprise’s content into information that is clean, rich, timely, consistent and structured.  The Xtractica® PCDM software system is the result of over $15 Million in software research and development. Xtractica® makes data from different sources coherent by resolving data inconsistencies and normalizing the data according to a uniform schema (ontology) or standard. The tool combines advanced automated tools for data acquisition, classification and cleaning.